E2E10 갤러리 — BGFIRST2 검증 + 결함 시연

Create ONE FINAL photorealistic live-action film still of the moment below — contemporary South Korea, 2026; all people are Korean unless stated. TIME OF DAY (lock): day.

SHOT TEXT (authoritative, Korean): 수리영이 강민숙의 어깨 쪽에 코를 묻은 채 눈을 감고 미소 짓는 옆모습

LOCATION (lock): Inside the rooftop apartment's entry area, where the daughter is putting on her shoes beside her mother. Daylight from the small living-room window reaches the home. The shot takes place here — the attached LOCATION PHOTOGRAPH shows the exact spot.

CAMERA & FRAME (follow exactly — this is the composition authority for this still; the reference images supply identity and place, never the framing):
- CAMERA: At shoulder height beside both women, the dolly-in settles into a close side composition perpendicular to 수리영's profile, placing her closed eyes and smile in the left-center as her nose rests near 강민숙's shoulder. 강민숙's shoulder and upper arm form the right foreground edge, leaving the lower frame open for the coming descent toward her wrist.
- FRAMING SCALE: close-up
- FRAME LAYOUT: 수리영 in the middle-left of the frame, midground; 강민숙 in the middle-right of the frame, foreground.
- KEY BACKGROUND ELEMENTS: 현관 공간 (수리영 is dressed to go out and putting on her shoes); used as Soft background context for the affectionate pause near the exit; 거실의 작은 창 (Letting in a thin line of sunlight) — Sunlight is visible entering through the window into the room; used as Secondary interior context behind the figures.
Compose the frame exactly as specified above — camera angle, subject scale and screen placement. Keep true physical scale between people and background elements: fixtures and distant objects occupy only the small screen area their real size and distance dictate; never enlarge a background object into a foreground presence.

REALIZE FIGURATIVE LANGUAGE AS A LIVE-ACTION SHOT: the Korean shot
text may describe characters metaphorically, figuratively or with
exaggeration. Photograph what a real movie camera would actually
record on a physical set — exaggerated or figurative impressions
become realistic staging choices (lighting, distance, angle,
wardrobe), not literal fantasy imagery.
EVERY CHARACTER IS A HUMAN BEING: unless the story explicitly
features non-human or virtual beings (as in science fiction or
fantasy), every character — however indirectly, vaguely or
figuratively the text describes them — IS a real human. When the
text gives no direct visual description of a person, IMAGINE one
and still show them as a concrete, fully-formed human being:
always render the human form to the maximum extent — build,
posture, face, hands, clothing — never reduce a person to a shape,
blob, solid silhouette or abstract mass.

EXPRESSIONS ARE ACTED, NEVER ANATOMICAL: when the text describes a
person's eyes, face or presence emotionally or figuratively (vacant,
hollow, dazed, burning, lifeless gaze and the like), it describes an
ACTOR'S PERFORMANCE captured by a real camera — realize it ONLY
through gaze direction, focus, eyelids, facial muscles, stillness
and posture. Every living person's eyes remain anatomically normal
human eyes with a natural iris and pupil, natural sclera and normal
proportions; NEVER whiten, blank out, cloud over, glow, enlarge or
otherwise alter eyeballs, skin or anatomy — unless the story
explicitly declares that being non-human or supernatural in form.

PROPS FACE THE RIGHT WAY: every handheld or used object must be
oriented exactly as its real-world use requires. A person reading,
watching or operating something (a phone, a photograph, a paper,
any device) has its functional side — screen, front, page — facing
THEIR OWN eyes; the camera then sees whatever side the staging
geometry implies (often its back). Show the functional side to the
camera ONLY when the shot text itself stages it toward the viewer.
Never flip, mirror or reverse an object's front and back.

CARRIED STATE (persist exactly — must match the neighbouring shots of this scene): Min-suk is still wearing her apron, and the small circular mark remains on her wrist even before Su-ri-young notices it.

PEOPLE: the SHOT TEXT alone decides whether any person is visible in this shot. IF a person appears, they must be one of: 강민숙 (한국인 40대 초반 여성, 단정한 검은 머리, 자연스러운 중년 얼굴); 수리영 (한국인 20대 초반 여성, 긴 검은 생머리, 젊은 타원형 얼굴) — never anyone else, and never add a person the shot text does not show. IF only part of a person is in frame (a hand, arm, foot, back, silhouette), that body part belongs to the specific person the shot text names — its sex, age, build, skin and grooming must unmistakably match that person's profile above.

No text, captions, watermarks or annotations anywhere.

Stage the shot. The FIRST attached image (SHOT BACKGROUND) is the finished empty background of this shot — keep it EXACTLY: its camera, perspective, architecture, lighting and every fixed feature stay untouched. The SECOND attached image (LAYOUT SKETCH) tells you ONLY where the people go: each sketched person's position, screen size, pose and the gaze/motion arrows. Ignore the sketch's background lines. The CHARACTER REFERENCE photographs show the real people.
Place the real people into the background at exactly the sketched positions, sizes and poses, following the arrow directions. No sketch lines or arrows may remain.

Create ONE FINAL photorealistic live-action film still of the moment below — contemporary South Korea, 2026; all people are Korean unless stated. TIME OF DAY (lock): day.

SHOT TEXT (authoritative, Korean): 수리영이 강민숙의 어깨 쪽에 코를 묻은 채 눈을 감고 미소 짓는 옆모습

LOCATION (lock): Inside the rooftop apartment's entry area, where the daughter is putting on her shoes beside her mother. Daylight from the small living-room window reaches the home. The shot takes place here — the attached LOCATION PHOTOGRAPH shows the exact spot.

CAMERA & FRAME (follow exactly — this is the composition authority for this still; the reference images supply identity and place, never the framing):
- CAMERA: At shoulder height beside both women, the dolly-in settles into a close side composition perpendicular to 수리영's profile, placing her closed eyes and smile in the left-center as her nose rests near 강민숙's shoulder. 강민숙's shoulder and upper arm form the right foreground edge, leaving the lower frame open for the coming descent toward her wrist.
- FRAMING SCALE: close-up
- FRAME LAYOUT: 수리영 in the middle-left of the frame, midground; 강민숙 in the middle-right of the frame, foreground.
- KEY BACKGROUND ELEMENTS: 현관 공간 (수리영 is dressed to go out and putting on her shoes); used as Soft background context for the affectionate pause near the exit; 거실의 작은 창 (Letting in a thin line of sunlight) — Sunlight is visible entering through the window into the room; used as Secondary interior context behind the figures.
Compose the frame exactly as specified above — camera angle, subject scale and screen placement. Keep true physical scale between people and background elements: fixtures and distant objects occupy only the small screen area their real size and distance dictate; never enlarge a background object into a foreground presence.

REALIZE FIGURATIVE LANGUAGE AS A LIVE-ACTION SHOT: the Korean shot
text may describe characters metaphorically, figuratively or with
exaggeration. Photograph what a real movie camera would actually
record on a physical set — exaggerated or figurative impressions
become realistic staging choices (lighting, distance, angle,
wardrobe), not literal fantasy imagery.
EVERY CHARACTER IS A HUMAN BEING: unless the story explicitly
features non-human or virtual beings (as in science fiction or
fantasy), every character — however indirectly, vaguely or
figuratively the text describes them — IS a real human. When the
text gives no direct visual description of a person, IMAGINE one
and still show them as a concrete, fully-formed human being:
always render the human form to the maximum extent — build,
posture, face, hands, clothing — never reduce a person to a shape,
blob, solid silhouette or abstract mass.

EXPRESSIONS ARE ACTED, NEVER ANATOMICAL: when the text describes a
person's eyes, face or presence emotionally or figuratively (vacant,
hollow, dazed, burning, lifeless gaze and the like), it describes an
ACTOR'S PERFORMANCE captured by a real camera — realize it ONLY
through gaze direction, focus, eyelids, facial muscles, stillness
and posture. Every living person's eyes remain anatomically normal
human eyes with a natural iris and pupil, natural sclera and normal
proportions; NEVER whiten, blank out, cloud over, glow, enlarge or
otherwise alter eyeballs, skin or anatomy — unless the story
explicitly declares that being non-human or supernatural in form.

PROPS FACE THE RIGHT WAY: every handheld or used object must be
oriented exactly as its real-world use requires. A person reading,
watching or operating something (a phone, a photograph, a paper,
any device) has its functional side — screen, front, page — facing
THEIR OWN eyes; the camera then sees whatever side the staging
geometry implies (often its back). Show the functional side to the
camera ONLY when the shot text itself stages it toward the viewer.
Never flip, mirror or reverse an object's front and back.

CARRIED STATE (persist exactly — must match the neighbouring shots of this scene): Min-suk is still wearing her apron, and the small circular mark remains on her wrist even before Su-ri-young notices it.

PEOPLE: the SHOT TEXT alone decides whether any person is visible in this shot. IF a person appears, they must be one of: 강민숙 (한국인 40대 초반 여성, 단정한 검은 머리, 자연스러운 중년 얼굴); 수리영 (한국인 20대 초반 여성, 긴 검은 생머리, 젊은 타원형 얼굴) — never anyone else, and never add a person the shot text does not show. IF only part of a person is in frame (a hand, arm, foot, back, silhouette), that body part belongs to the specific person the shot text names — its sex, age, build, skin and grooming must unmistakably match that person's profile above.

No text, captions, watermarks or annotations anywhere.

Create ONE FINAL photorealistic live-action film still of the moment below — contemporary South Korea, 2026; all people are Korean unless stated. TIME OF DAY (lock): day.

SHOT TEXT (authoritative, Korean): 수리영이 강민숙의 어깨 쪽에 코를 묻은 채 눈을 감고 미소 짓는 옆모습

LOCATION (lock): Inside the rooftop apartment's entry area, where the daughter is putting on her shoes beside her mother. Daylight from the small living-room window reaches the home. The shot takes place here — the attached LOCATION PHOTOGRAPH shows the exact spot.

CAMERA & FRAME (follow exactly — this is the composition authority for this still; the reference images supply identity and place, never the framing):
- CAMERA: At shoulder height beside both women, the dolly-in settles into a close side composition perpendicular to 수리영's profile, placing her closed eyes and smile in the left-center as her nose rests near 강민숙's shoulder. 강민숙's shoulder and upper arm form the right foreground edge, leaving the lower frame open for the coming descent toward her wrist.
- FRAMING SCALE: close-up
- FRAME LAYOUT: 수리영 in the middle-left of the frame, midground; 강민숙 in the middle-right of the frame, foreground.
- KEY BACKGROUND ELEMENTS: 현관 공간 (수리영 is dressed to go out and putting on her shoes); used as Soft background context for the affectionate pause near the exit; 거실의 작은 창 (Letting in a thin line of sunlight) — Sunlight is visible entering through the window into the room; used as Secondary interior context behind the figures.
Compose the frame exactly as specified above — camera angle, subject scale and screen placement. Keep true physical scale between people and background elements: fixtures and distant objects occupy only the small screen area their real size and distance dictate; never enlarge a background object into a foreground presence.

REALIZE FIGURATIVE LANGUAGE AS A LIVE-ACTION SHOT: the Korean shot
text may describe characters metaphorically, figuratively or with
exaggeration. Photograph what a real movie camera would actually
record on a physical set — exaggerated or figurative impressions
become realistic staging choices (lighting, distance, angle,
wardrobe), not literal fantasy imagery.
EVERY CHARACTER IS A HUMAN BEING: unless the story explicitly
features non-human or virtual beings (as in science fiction or
fantasy), every character — however indirectly, vaguely or
figuratively the text describes them — IS a real human. When the
text gives no direct visual description of a person, IMAGINE one
and still show them as a concrete, fully-formed human being:
always render the human form to the maximum extent — build,
posture, face, hands, clothing — never reduce a person to a shape,
blob, solid silhouette or abstract mass.

EXPRESSIONS ARE ACTED, NEVER ANATOMICAL: when the text describes a
person's eyes, face or presence emotionally or figuratively (vacant,
hollow, dazed, burning, lifeless gaze and the like), it describes an
ACTOR'S PERFORMANCE captured by a real camera — realize it ONLY
through gaze direction, focus, eyelids, facial muscles, stillness
and posture. Every living person's eyes remain anatomically normal
human eyes with a natural iris and pupil, natural sclera and normal
proportions; NEVER whiten, blank out, cloud over, glow, enlarge or
otherwise alter eyeballs, skin or anatomy — unless the story
explicitly declares that being non-human or supernatural in form.

PROPS FACE THE RIGHT WAY: every handheld or used object must be
oriented exactly as its real-world use requires. A person reading,
watching or operating something (a phone, a photograph, a paper,
any device) has its functional side — screen, front, page — facing
THEIR OWN eyes; the camera then sees whatever side the staging
geometry implies (often its back). Show the functional side to the
camera ONLY when the shot text itself stages it toward the viewer.
Never flip, mirror or reverse an object's front and back.

CARRIED STATE (persist exactly — must match the neighbouring shots of this scene): Min-suk is still wearing her apron, and the small circular mark remains on her wrist even before Su-ri-young notices it.

PEOPLE: the SHOT TEXT alone decides whether any person is visible in this shot. IF a person appears, they must be one of: 강민숙 (한국인 40대 초반 여성, 단정한 검은 머리, 자연스러운 중년 얼굴); 수리영 (한국인 20대 초반 여성, 긴 검은 생머리, 젊은 타원형 얼굴) — never anyone else, and never add a person the shot text does not show. IF only part of a person is in frame (a hand, arm, foot, back, silhouette), that body part belongs to the specific person the shot text names — its sex, age, build, skin and grooming must unmistakably match that person's profile above.

No text, captions, watermarks or annotations anywhere.

Edit this photograph. Apply ONLY the corrections listed below,
each in place, changing nothing else:

CORRECTIONS:
- Add the dark backpack strap over Su-ri-young's shoulder and chest to match her character reference.

PRESERVE EVERYTHING ELSE EXACTLY as in the original photograph —
every person, pose, face, garment, object, set feature, lighting,
camera and framing stay identical. No text, captions or
watermarks anywhere.

Create ONE FINAL photorealistic live-action film still of the moment below — contemporary South Korea, 2026; all people are Korean unless stated. TIME OF DAY (lock): dusk.

SHOT TEXT (authoritative, Korean): 수리영이 눈을 동그랗게 뜬 채 의아한 표정으로 서 있는 얼굴 클로즈업

LOCATION (lock): Inside the large supermarket on the merchandise floor, beside employees stocking the displays under store lighting. The shot takes place here — the attached LOCATION PHOTOGRAPH shows the exact spot.

CAMERA & FRAME (follow exactly — this is the composition authority for this still; the reference images supply identity and place, never the framing):
- CAMERA: From inside the store on the employees’ side, an eye-level dolly-in resolves into a tight close-up of 수리영, with her widened eyes held near the upper-center and her tense mouth in the lower third. Her face is seen at a three-quarter angle as the store shelving falls softly out of focus, emphasizing only the reduced camera distance at the end of the inward move.
- FRAMING SCALE: close-up
- FRAME LAYOUT: 수리영 in the middle-center of the frame, foreground.
- KEY BACKGROUND ELEMENTS: 마트 진열대 (Present in the store; no further state specified) — A partial aisle-facing side recedes behind 수리영; used as Softly defocused context behind the close facial framing.
Compose the frame exactly as specified above — camera angle, subject scale and screen placement. Keep true physical scale between people and background elements: fixtures and distant objects occupy only the small screen area their real size and distance dictate; never enlarge a background object into a foreground presence.

REALIZE FIGURATIVE LANGUAGE AS A LIVE-ACTION SHOT: the Korean shot
text may describe characters metaphorically, figuratively or with
exaggeration. Photograph what a real movie camera would actually
record on a physical set — exaggerated or figurative impressions
become realistic staging choices (lighting, distance, angle,
wardrobe), not literal fantasy imagery.
EVERY CHARACTER IS A HUMAN BEING: unless the story explicitly
features non-human or virtual beings (as in science fiction or
fantasy), every character — however indirectly, vaguely or
figuratively the text describes them — IS a real human. When the
text gives no direct visual description of a person, IMAGINE one
and still show them as a concrete, fully-formed human being:
always render the human form to the maximum extent — build,
posture, face, hands, clothing — never reduce a person to a shape,
blob, solid silhouette or abstract mass.

EXPRESSIONS ARE ACTED, NEVER ANATOMICAL: when the text describes a
person's eyes, face or presence emotionally or figuratively (vacant,
hollow, dazed, burning, lifeless gaze and the like), it describes an
ACTOR'S PERFORMANCE captured by a real camera — realize it ONLY
through gaze direction, focus, eyelids, facial muscles, stillness
and posture. Every living person's eyes remain anatomically normal
human eyes with a natural iris and pupil, natural sclera and normal
proportions; NEVER whiten, blank out, cloud over, glow, enlarge or
otherwise alter eyeballs, skin or anatomy — unless the story
explicitly declares that being non-human or supernatural in form.

PROPS FACE THE RIGHT WAY: every handheld or used object must be
oriented exactly as its real-world use requires. A person reading,
watching or operating something (a phone, a photograph, a paper,
any device) has its functional side — screen, front, page — facing
THEIR OWN eyes; the camera then sees whatever side the staging
geometry implies (often its back). Show the functional side to the
camera ONLY when the shot text itself stages it toward the viewer.
Never flip, mirror or reverse an object's front and back.

CARRIED STATE (persist exactly — must match the neighbouring shots of this scene): Su-ri-young’s bicycle remains parked outside the mart, and she still has the phone she used to contact Min-suk.

PEOPLE: the SHOT TEXT alone decides whether any person is visible in this shot. IF a person appears, they must be one of: 수리영 (한국인 20대 초반 여성, 긴 검은 생머리, 젊은 타원형 얼굴) — never anyone else, and never add a person the shot text does not show. IF only part of a person is in frame (a hand, arm, foot, back, silhouette), that body part belongs to the specific person the shot text names — its sex, age, build, skin and grooming must unmistakably match that person's profile above.

No text, captions, watermarks or annotations anywhere.

Stage the shot. The FIRST attached image (SHOT BACKGROUND) is the finished empty background of this shot — keep it EXACTLY: its camera, perspective, architecture, lighting and every fixed feature stay untouched. The SECOND attached image (LAYOUT SKETCH) tells you ONLY where the people go: each sketched person's position, screen size, pose and the gaze/motion arrows. Ignore the sketch's background lines. The CHARACTER REFERENCE photographs show the real people.
Place the real people into the background at exactly the sketched positions, sizes and poses, following the arrow directions. No sketch lines or arrows may remain.

Create ONE FINAL photorealistic live-action film still of the moment below — contemporary South Korea, 2026; all people are Korean unless stated. TIME OF DAY (lock): dusk.

SHOT TEXT (authoritative, Korean): 수리영이 눈을 동그랗게 뜬 채 의아한 표정으로 서 있는 얼굴 클로즈업

LOCATION (lock): Inside the large supermarket on the merchandise floor, beside employees stocking the displays under store lighting. The shot takes place here — the attached LOCATION PHOTOGRAPH shows the exact spot.

CAMERA & FRAME (follow exactly — this is the composition authority for this still; the reference images supply identity and place, never the framing):
- CAMERA: From inside the store on the employees’ side, an eye-level dolly-in resolves into a tight close-up of 수리영, with her widened eyes held near the upper-center and her tense mouth in the lower third. Her face is seen at a three-quarter angle as the store shelving falls softly out of focus, emphasizing only the reduced camera distance at the end of the inward move.
- FRAMING SCALE: close-up
- FRAME LAYOUT: 수리영 in the middle-center of the frame, foreground.
- KEY BACKGROUND ELEMENTS: 마트 진열대 (Present in the store; no further state specified) — A partial aisle-facing side recedes behind 수리영; used as Softly defocused context behind the close facial framing.
Compose the frame exactly as specified above — camera angle, subject scale and screen placement. Keep true physical scale between people and background elements: fixtures and distant objects occupy only the small screen area their real size and distance dictate; never enlarge a background object into a foreground presence.

REALIZE FIGURATIVE LANGUAGE AS A LIVE-ACTION SHOT: the Korean shot
text may describe characters metaphorically, figuratively or with
exaggeration. Photograph what a real movie camera would actually
record on a physical set — exaggerated or figurative impressions
become realistic staging choices (lighting, distance, angle,
wardrobe), not literal fantasy imagery.
EVERY CHARACTER IS A HUMAN BEING: unless the story explicitly
features non-human or virtual beings (as in science fiction or
fantasy), every character — however indirectly, vaguely or
figuratively the text describes them — IS a real human. When the
text gives no direct visual description of a person, IMAGINE one
and still show them as a concrete, fully-formed human being:
always render the human form to the maximum extent — build,
posture, face, hands, clothing — never reduce a person to a shape,
blob, solid silhouette or abstract mass.

EXPRESSIONS ARE ACTED, NEVER ANATOMICAL: when the text describes a
person's eyes, face or presence emotionally or figuratively (vacant,
hollow, dazed, burning, lifeless gaze and the like), it describes an
ACTOR'S PERFORMANCE captured by a real camera — realize it ONLY
through gaze direction, focus, eyelids, facial muscles, stillness
and posture. Every living person's eyes remain anatomically normal
human eyes with a natural iris and pupil, natural sclera and normal
proportions; NEVER whiten, blank out, cloud over, glow, enlarge or
otherwise alter eyeballs, skin or anatomy — unless the story
explicitly declares that being non-human or supernatural in form.

PROPS FACE THE RIGHT WAY: every handheld or used object must be
oriented exactly as its real-world use requires. A person reading,
watching or operating something (a phone, a photograph, a paper,
any device) has its functional side — screen, front, page — facing
THEIR OWN eyes; the camera then sees whatever side the staging
geometry implies (often its back). Show the functional side to the
camera ONLY when the shot text itself stages it toward the viewer.
Never flip, mirror or reverse an object's front and back.

CARRIED STATE (persist exactly — must match the neighbouring shots of this scene): Su-ri-young’s bicycle remains parked outside the mart, and she still has the phone she used to contact Min-suk.

PEOPLE: the SHOT TEXT alone decides whether any person is visible in this shot. IF a person appears, they must be one of: 수리영 (한국인 20대 초반 여성, 긴 검은 생머리, 젊은 타원형 얼굴) — never anyone else, and never add a person the shot text does not show. IF only part of a person is in frame (a hand, arm, foot, back, silhouette), that body part belongs to the specific person the shot text names — its sex, age, build, skin and grooming must unmistakably match that person's profile above.

No text, captions, watermarks or annotations anywhere.

Create ONE FINAL photorealistic live-action film still of the moment below — contemporary South Korea, 2026; all people are Korean unless stated. TIME OF DAY (lock): dusk.

SHOT TEXT (authoritative, Korean): 수리영이 눈을 동그랗게 뜬 채 의아한 표정으로 서 있는 얼굴 클로즈업

LOCATION (lock): Inside the large supermarket on the merchandise floor, beside employees stocking the displays under store lighting. The shot takes place here — the attached LOCATION PHOTOGRAPH shows the exact spot.

CAMERA & FRAME (follow exactly — this is the composition authority for this still; the reference images supply identity and place, never the framing):
- CAMERA: From inside the store on the employees’ side, an eye-level dolly-in resolves into a tight close-up of 수리영, with her widened eyes held near the upper-center and her tense mouth in the lower third. Her face is seen at a three-quarter angle as the store shelving falls softly out of focus, emphasizing only the reduced camera distance at the end of the inward move.
- FRAMING SCALE: close-up
- FRAME LAYOUT: 수리영 in the middle-center of the frame, foreground.
- KEY BACKGROUND ELEMENTS: 마트 진열대 (Present in the store; no further state specified) — A partial aisle-facing side recedes behind 수리영; used as Softly defocused context behind the close facial framing.
Compose the frame exactly as specified above — camera angle, subject scale and screen placement. Keep true physical scale between people and background elements: fixtures and distant objects occupy only the small screen area their real size and distance dictate; never enlarge a background object into a foreground presence.

REALIZE FIGURATIVE LANGUAGE AS A LIVE-ACTION SHOT: the Korean shot
text may describe characters metaphorically, figuratively or with
exaggeration. Photograph what a real movie camera would actually
record on a physical set — exaggerated or figurative impressions
become realistic staging choices (lighting, distance, angle,
wardrobe), not literal fantasy imagery.
EVERY CHARACTER IS A HUMAN BEING: unless the story explicitly
features non-human or virtual beings (as in science fiction or
fantasy), every character — however indirectly, vaguely or
figuratively the text describes them — IS a real human. When the
text gives no direct visual description of a person, IMAGINE one
and still show them as a concrete, fully-formed human being:
always render the human form to the maximum extent — build,
posture, face, hands, clothing — never reduce a person to a shape,
blob, solid silhouette or abstract mass.

EXPRESSIONS ARE ACTED, NEVER ANATOMICAL: when the text describes a
person's eyes, face or presence emotionally or figuratively (vacant,
hollow, dazed, burning, lifeless gaze and the like), it describes an
ACTOR'S PERFORMANCE captured by a real camera — realize it ONLY
through gaze direction, focus, eyelids, facial muscles, stillness
and posture. Every living person's eyes remain anatomically normal
human eyes with a natural iris and pupil, natural sclera and normal
proportions; NEVER whiten, blank out, cloud over, glow, enlarge or
otherwise alter eyeballs, skin or anatomy — unless the story
explicitly declares that being non-human or supernatural in form.

PROPS FACE THE RIGHT WAY: every handheld or used object must be
oriented exactly as its real-world use requires. A person reading,
watching or operating something (a phone, a photograph, a paper,
any device) has its functional side — screen, front, page — facing
THEIR OWN eyes; the camera then sees whatever side the staging
geometry implies (often its back). Show the functional side to the
camera ONLY when the shot text itself stages it toward the viewer.
Never flip, mirror or reverse an object's front and back.

CARRIED STATE (persist exactly — must match the neighbouring shots of this scene): Su-ri-young’s bicycle remains parked outside the mart, and she still has the phone she used to contact Min-suk.

PEOPLE: the SHOT TEXT alone decides whether any person is visible in this shot. IF a person appears, they must be one of: 수리영 (한국인 20대 초반 여성, 긴 검은 생머리, 젊은 타원형 얼굴) — never anyone else, and never add a person the shot text does not show. IF only part of a person is in frame (a hand, arm, foot, back, silhouette), that body part belongs to the specific person the shot text names — its sex, age, build, skin and grooming must unmistakably match that person's profile above.

No text, captions, watermarks or annotations anywhere.

Edit this photograph. Apply ONLY the corrections listed below,
each in place, changing nothing else:

CORRECTIONS:
- Turn the character's head so her face is seen at a three-quarter angle instead of straight on.

PRESERVE EVERYTHING ELSE EXACTLY as in the original photograph —
every person, pose, face, garment, object, set feature, lighting,
camera and framing stay identical. No text, captions or
watermarks anywhere.

Create ONE FINAL photorealistic live-action film still of the moment below — contemporary South Korea, 2026; all people are Korean unless stated. TIME OF DAY (lock): early morning.

SHOT TEXT (authoritative, Korean): 마트 밖, 수리영과 마트 남자 직원이 나란히 서서 건물 외벽의 CCTV 카메라를 올려다보는 뒷모습

LOCATION (lock): Outside the supermarket, directly beneath the section of exterior wall fitted with a CCTV camera. The shot takes place here — the attached LOCATION PHOTOGRAPH shows the exact spot.

CAMERA & FRAME (follow exactly — this is the composition authority for this still; the reference images supply identity and place, never the framing):
- CAMERA: From eye level several meters behind them, hold a static wide view of 수리영 and 마트 남자 직원 side by side, both backs to camera as they tilt their heads toward the CCTV camera above on the exterior wall. Keep their full figures in the lower half and the surveillance camera high within the same frame, establishing a clean vertical sightline for the upcoming tilt.
- FRAMING SCALE: wide shot
- FRAME LAYOUT: 수리영 in the lower-left of the frame, midground, looks toward exterior CCTV camera; 마트 남자 직원 in the lower-right of the frame, midground, looks toward exterior CCTV camera; exterior CCTV camera in the upper-center of the frame, background.
- KEY BACKGROUND ELEMENTS: exterior CCTV camera (Mounted outside the mart) — Its lens side points toward the area the pair are checking, while the camera sees it from below and behind them; used as Provides the elevated target shared by both characters and the endpoint for the next upward tilt; mart exterior wall (Visible in the early morning) — The wall face carrying the CCTV camera is visible beyond the two figures; used as Establishes the exterior location and supports the CCTV camera's position above the pair.
Compose the frame exactly as specified above — camera angle, subject scale and screen placement. Keep true physical scale between people and background elements: fixtures and distant objects occupy only the small screen area their real size and distance dictate; never enlarge a background object into a foreground presence.

REALIZE FIGURATIVE LANGUAGE AS A LIVE-ACTION SHOT: the Korean shot
text may describe characters metaphorically, figuratively or with
exaggeration. Photograph what a real movie camera would actually
record on a physical set — exaggerated or figurative impressions
become realistic staging choices (lighting, distance, angle,
wardrobe), not literal fantasy imagery.
EVERY CHARACTER IS A HUMAN BEING: unless the story explicitly
features non-human or virtual beings (as in science fiction or
fantasy), every character — however indirectly, vaguely or
figuratively the text describes them — IS a real human. When the
text gives no direct visual description of a person, IMAGINE one
and still show them as a concrete, fully-formed human being:
always render the human form to the maximum extent — build,
posture, face, hands, clothing — never reduce a person to a shape,
blob, solid silhouette or abstract mass.

EXPRESSIONS ARE ACTED, NEVER ANATOMICAL: when the text describes a
person's eyes, face or presence emotionally or figuratively (vacant,
hollow, dazed, burning, lifeless gaze and the like), it describes an
ACTOR'S PERFORMANCE captured by a real camera — realize it ONLY
through gaze direction, focus, eyelids, facial muscles, stillness
and posture. Every living person's eyes remain anatomically normal
human eyes with a natural iris and pupil, natural sclera and normal
proportions; NEVER whiten, blank out, cloud over, glow, enlarge or
otherwise alter eyeballs, skin or anatomy — unless the story
explicitly declares that being non-human or supernatural in form.

PROPS FACE THE RIGHT WAY: every handheld or used object must be
oriented exactly as its real-world use requires. A person reading,
watching or operating something (a phone, a photograph, a paper,
any device) has its functional side — screen, front, page — facing
THEIR OWN eyes; the camera then sees whatever side the staging
geometry implies (often its back). Show the functional side to the
camera ONLY when the shot text itself stages it toward the viewer.
Never flip, mirror or reverse an object's front and back.

CARRIED STATE (persist exactly — must match the neighbouring shots of this scene): Su-ri-young’s bicycle remains parked at the mart, and she has brought her doll-adorned backpack and phone with her.

PEOPLE: the SHOT TEXT alone decides whether any person is visible in this shot. IF a person appears, they must be one of: 수리영 (한국인 20대 초반 여성, 긴 검은 생머리, 젊은 타원형 얼굴); 마트 남자 직원 (한국인 성인 남성, 짧은 검은 머리, 평범한 얼굴형) — never anyone else, and never add a person the shot text does not show. IF only part of a person is in frame (a hand, arm, foot, back, silhouette), that body part belongs to the specific person the shot text names — its sex, age, build, skin and grooming must unmistakably match that person's profile above.

No text, captions, watermarks or annotations anywhere.

Stage the shot. The FIRST attached image (SHOT BACKGROUND) is the finished empty background of this shot — keep it EXACTLY: its camera, perspective, architecture, lighting and every fixed feature stay untouched. The SECOND attached image (LAYOUT SKETCH) tells you ONLY where the people go: each sketched person's position, screen size, pose and the gaze/motion arrows. Ignore the sketch's background lines. The CHARACTER REFERENCE photographs show the real people.
Place the real people into the background at exactly the sketched positions, sizes and poses, following the arrow directions. No sketch lines or arrows may remain.

Create ONE FINAL photorealistic live-action film still of the moment below — contemporary South Korea, 2026; all people are Korean unless stated. TIME OF DAY (lock): early morning.

SHOT TEXT (authoritative, Korean): 마트 밖, 수리영과 마트 남자 직원이 나란히 서서 건물 외벽의 CCTV 카메라를 올려다보는 뒷모습

LOCATION (lock): Outside the supermarket, directly beneath the section of exterior wall fitted with a CCTV camera. The shot takes place here — the attached LOCATION PHOTOGRAPH shows the exact spot.

CAMERA & FRAME (follow exactly — this is the composition authority for this still; the reference images supply identity and place, never the framing):
- CAMERA: From eye level several meters behind them, hold a static wide view of 수리영 and 마트 남자 직원 side by side, both backs to camera as they tilt their heads toward the CCTV camera above on the exterior wall. Keep their full figures in the lower half and the surveillance camera high within the same frame, establishing a clean vertical sightline for the upcoming tilt.
- FRAMING SCALE: wide shot
- FRAME LAYOUT: 수리영 in the lower-left of the frame, midground, looks toward exterior CCTV camera; 마트 남자 직원 in the lower-right of the frame, midground, looks toward exterior CCTV camera; exterior CCTV camera in the upper-center of the frame, background.
- KEY BACKGROUND ELEMENTS: exterior CCTV camera (Mounted outside the mart) — Its lens side points toward the area the pair are checking, while the camera sees it from below and behind them; used as Provides the elevated target shared by both characters and the endpoint for the next upward tilt; mart exterior wall (Visible in the early morning) — The wall face carrying the CCTV camera is visible beyond the two figures; used as Establishes the exterior location and supports the CCTV camera's position above the pair.
Compose the frame exactly as specified above — camera angle, subject scale and screen placement. Keep true physical scale between people and background elements: fixtures and distant objects occupy only the small screen area their real size and distance dictate; never enlarge a background object into a foreground presence.

REALIZE FIGURATIVE LANGUAGE AS A LIVE-ACTION SHOT: the Korean shot
text may describe characters metaphorically, figuratively or with
exaggeration. Photograph what a real movie camera would actually
record on a physical set — exaggerated or figurative impressions
become realistic staging choices (lighting, distance, angle,
wardrobe), not literal fantasy imagery.
EVERY CHARACTER IS A HUMAN BEING: unless the story explicitly
features non-human or virtual beings (as in science fiction or
fantasy), every character — however indirectly, vaguely or
figuratively the text describes them — IS a real human. When the
text gives no direct visual description of a person, IMAGINE one
and still show them as a concrete, fully-formed human being:
always render the human form to the maximum extent — build,
posture, face, hands, clothing — never reduce a person to a shape,
blob, solid silhouette or abstract mass.

EXPRESSIONS ARE ACTED, NEVER ANATOMICAL: when the text describes a
person's eyes, face or presence emotionally or figuratively (vacant,
hollow, dazed, burning, lifeless gaze and the like), it describes an
ACTOR'S PERFORMANCE captured by a real camera — realize it ONLY
through gaze direction, focus, eyelids, facial muscles, stillness
and posture. Every living person's eyes remain anatomically normal
human eyes with a natural iris and pupil, natural sclera and normal
proportions; NEVER whiten, blank out, cloud over, glow, enlarge or
otherwise alter eyeballs, skin or anatomy — unless the story
explicitly declares that being non-human or supernatural in form.

PROPS FACE THE RIGHT WAY: every handheld or used object must be
oriented exactly as its real-world use requires. A person reading,
watching or operating something (a phone, a photograph, a paper,
any device) has its functional side — screen, front, page — facing
THEIR OWN eyes; the camera then sees whatever side the staging
geometry implies (often its back). Show the functional side to the
camera ONLY when the shot text itself stages it toward the viewer.
Never flip, mirror or reverse an object's front and back.

CARRIED STATE (persist exactly — must match the neighbouring shots of this scene): Su-ri-young’s bicycle remains parked at the mart, and she has brought her doll-adorned backpack and phone with her.

PEOPLE: the SHOT TEXT alone decides whether any person is visible in this shot. IF a person appears, they must be one of: 수리영 (한국인 20대 초반 여성, 긴 검은 생머리, 젊은 타원형 얼굴); 마트 남자 직원 (한국인 성인 남성, 짧은 검은 머리, 평범한 얼굴형) — never anyone else, and never add a person the shot text does not show. IF only part of a person is in frame (a hand, arm, foot, back, silhouette), that body part belongs to the specific person the shot text names — its sex, age, build, skin and grooming must unmistakably match that person's profile above.

No text, captions, watermarks or annotations anywhere.

Create ONE FINAL photorealistic live-action film still of the moment below — contemporary South Korea, 2026; all people are Korean unless stated. TIME OF DAY (lock): early morning.

SHOT TEXT (authoritative, Korean): 마트 밖, 수리영과 마트 남자 직원이 나란히 서서 건물 외벽의 CCTV 카메라를 올려다보는 뒷모습

LOCATION (lock): Outside the supermarket, directly beneath the section of exterior wall fitted with a CCTV camera. The shot takes place here — the attached LOCATION PHOTOGRAPH shows the exact spot.

CAMERA & FRAME (follow exactly — this is the composition authority for this still; the reference images supply identity and place, never the framing):
- CAMERA: From eye level several meters behind them, hold a static wide view of 수리영 and 마트 남자 직원 side by side, both backs to camera as they tilt their heads toward the CCTV camera above on the exterior wall. Keep their full figures in the lower half and the surveillance camera high within the same frame, establishing a clean vertical sightline for the upcoming tilt.
- FRAMING SCALE: wide shot
- FRAME LAYOUT: 수리영 in the lower-left of the frame, midground, looks toward exterior CCTV camera; 마트 남자 직원 in the lower-right of the frame, midground, looks toward exterior CCTV camera; exterior CCTV camera in the upper-center of the frame, background.
- KEY BACKGROUND ELEMENTS: exterior CCTV camera (Mounted outside the mart) — Its lens side points toward the area the pair are checking, while the camera sees it from below and behind them; used as Provides the elevated target shared by both characters and the endpoint for the next upward tilt; mart exterior wall (Visible in the early morning) — The wall face carrying the CCTV camera is visible beyond the two figures; used as Establishes the exterior location and supports the CCTV camera's position above the pair.
Compose the frame exactly as specified above — camera angle, subject scale and screen placement. Keep true physical scale between people and background elements: fixtures and distant objects occupy only the small screen area their real size and distance dictate; never enlarge a background object into a foreground presence.

REALIZE FIGURATIVE LANGUAGE AS A LIVE-ACTION SHOT: the Korean shot
text may describe characters metaphorically, figuratively or with
exaggeration. Photograph what a real movie camera would actually
record on a physical set — exaggerated or figurative impressions
become realistic staging choices (lighting, distance, angle,
wardrobe), not literal fantasy imagery.
EVERY CHARACTER IS A HUMAN BEING: unless the story explicitly
features non-human or virtual beings (as in science fiction or
fantasy), every character — however indirectly, vaguely or
figuratively the text describes them — IS a real human. When the
text gives no direct visual description of a person, IMAGINE one
and still show them as a concrete, fully-formed human being:
always render the human form to the maximum extent — build,
posture, face, hands, clothing — never reduce a person to a shape,
blob, solid silhouette or abstract mass.

EXPRESSIONS ARE ACTED, NEVER ANATOMICAL: when the text describes a
person's eyes, face or presence emotionally or figuratively (vacant,
hollow, dazed, burning, lifeless gaze and the like), it describes an
ACTOR'S PERFORMANCE captured by a real camera — realize it ONLY
through gaze direction, focus, eyelids, facial muscles, stillness
and posture. Every living person's eyes remain anatomically normal
human eyes with a natural iris and pupil, natural sclera and normal
proportions; NEVER whiten, blank out, cloud over, glow, enlarge or
otherwise alter eyeballs, skin or anatomy — unless the story
explicitly declares that being non-human or supernatural in form.

PROPS FACE THE RIGHT WAY: every handheld or used object must be
oriented exactly as its real-world use requires. A person reading,
watching or operating something (a phone, a photograph, a paper,
any device) has its functional side — screen, front, page — facing
THEIR OWN eyes; the camera then sees whatever side the staging
geometry implies (often its back). Show the functional side to the
camera ONLY when the shot text itself stages it toward the viewer.
Never flip, mirror or reverse an object's front and back.

CARRIED STATE (persist exactly — must match the neighbouring shots of this scene): Su-ri-young’s bicycle remains parked at the mart, and she has brought her doll-adorned backpack and phone with her.

PEOPLE: the SHOT TEXT alone decides whether any person is visible in this shot. IF a person appears, they must be one of: 수리영 (한국인 20대 초반 여성, 긴 검은 생머리, 젊은 타원형 얼굴); 마트 남자 직원 (한국인 성인 남성, 짧은 검은 머리, 평범한 얼굴형) — never anyone else, and never add a person the shot text does not show. IF only part of a person is in frame (a hand, arm, foot, back, silhouette), that body part belongs to the specific person the shot text names — its sex, age, build, skin and grooming must unmistakably match that person's profile above.

No text, captions, watermarks or annotations anywhere.

Edit this photograph. Apply ONLY the corrections listed below,
each in place, changing nothing else:

CORRECTIONS:
- Add the dark green cap from the character reference to Su-ri-young's head.
- Move the parked bicycle to the left side of Su-ri-young, near the AC units, as indicated in the layout sketch.

PRESERVE EVERYTHING ELSE EXACTLY as in the original photograph —
every person, pose, face, garment, object, set feature, lighting,
camera and framing stay identical. No text, captions or
watermarks anywhere.

Create ONE FINAL photorealistic live-action film still of the moment below — contemporary South Korea, 2026; all people are Korean unless stated. TIME OF DAY (lock): early morning.

SHOT TEXT (authoritative, Korean): 책상 앞 모니터를 향해 앉은 40대 한국인 남성 보안직원과 혜수의 뒷모습, 그 뒤에 선 수리영의 굳은 정면 구도

LOCATION (lock): Inside the supermarket's security office, at a desk with a CCTV monitor under office lighting. The shot takes place here — the attached LOCATION PHOTOGRAPH shows the exact spot.

CAMERA & FRAME (follow exactly — this is the composition authority for this still; the reference images supply identity and place, never the framing):
- CAMERA: Hold a static medium-wide composition from seated eye level behind the security employee and Hyesu, placing their backs across the lower left and right while the monitor sits between their shoulder lines. Suriyoung remains rigidly frontal in the central background, creating a compressed line of monitor, seated observers, and anxious witness before the later forward move.
- FRAMING SCALE: wide shot
- FRAME LAYOUT: 40대 한국인 남성 보안직원 in the lower-left of the frame, foreground; 혜수 in the lower-right of the frame, foreground; 수리영 in the upper-center of the frame, background, looks toward CCTV monitor.
- KEY BACKGROUND ELEMENTS: CCTV 모니터 (On and replaying footage) — Its active front face is visible, displaying CCTV playback with one section markedly darker and difficult to identify; used as Central visual anchor framed between the two seated figures; 책상 (In use) — The monitor-facing edge runs across the frame between the seated figures and Suriyoung; used as Supports the monitor and establishes the office work area across the midground; 의자 두 개 (Occupied) — The backs face the camera while the seats face the monitor; used as Forms two foreground blocks that compress Suriyoung in depth.
Compose the frame exactly as specified above — camera angle, subject scale and screen placement. Keep true physical scale between people and background elements: fixtures and distant objects occupy only the small screen area their real size and distance dictate; never enlarge a background object into a foreground presence.

REALIZE FIGURATIVE LANGUAGE AS A LIVE-ACTION SHOT: the Korean shot
text may describe characters metaphorically, figuratively or with
exaggeration. Photograph what a real movie camera would actually
record on a physical set — exaggerated or figurative impressions
become realistic staging choices (lighting, distance, angle,
wardrobe), not literal fantasy imagery.
EVERY CHARACTER IS A HUMAN BEING: unless the story explicitly
features non-human or virtual beings (as in science fiction or
fantasy), every character — however indirectly, vaguely or
figuratively the text describes them — IS a real human. When the
text gives no direct visual description of a person, IMAGINE one
and still show them as a concrete, fully-formed human being:
always render the human form to the maximum extent — build,
posture, face, hands, clothing — never reduce a person to a shape,
blob, solid silhouette or abstract mass.

EXPRESSIONS ARE ACTED, NEVER ANATOMICAL: when the text describes a
person's eyes, face or presence emotionally or figuratively (vacant,
hollow, dazed, burning, lifeless gaze and the like), it describes an
ACTOR'S PERFORMANCE captured by a real camera — realize it ONLY
through gaze direction, focus, eyelids, facial muscles, stillness
and posture. Every living person's eyes remain anatomically normal
human eyes with a natural iris and pupil, natural sclera and normal
proportions; NEVER whiten, blank out, cloud over, glow, enlarge or
otherwise alter eyeballs, skin or anatomy — unless the story
explicitly declares that being non-human or supernatural in form.

PROPS FACE THE RIGHT WAY: every handheld or used object must be
oriented exactly as its real-world use requires. A person reading,
watching or operating something (a phone, a photograph, a paper,
any device) has its functional side — screen, front, page — facing
THEIR OWN eyes; the camera then sees whatever side the staging
geometry implies (often its back). Show the functional side to the
camera ONLY when the shot text itself stages it toward the viewer.
Never flip, mirror or reverse an object's front and back.

CARRIED STATE (persist exactly — must match the neighbouring shots of this scene): The relevant section of the mart CCTV recording remains uniquely dark and impossible to identify through repeated playback. Su-ri-young still has her doll-adorned backpack with her.

PEOPLE: the SHOT TEXT alone decides whether any person is visible in this shot. IF a person appears, they must be one of: 혜수 (한국인 40대 초반 여성, 단정한 검은 머리, 성숙한 타원형 얼굴); 수리영 (한국인 20대 초반 여성, 긴 검은 생머리, 젊은 타원형 얼굴); 40대 한국인 남성 보안직원 (한국인 40대 남성, 짧은 검은 머리, 단정한 머리 모양, 중년 남성 얼굴) — never anyone else, and never add a person the shot text does not show. IF only part of a person is in frame (a hand, arm, foot, back, silhouette), that body part belongs to the specific person the shot text names — its sex, age, build, skin and grooming must unmistakably match that person's profile above.

No text, captions, watermarks or annotations anywhere.

Stage the shot. The FIRST attached image (SHOT BACKGROUND) is the finished empty background of this shot — keep it EXACTLY: its camera, perspective, architecture, lighting and every fixed feature stay untouched. The SECOND attached image (LAYOUT SKETCH) tells you ONLY where the people go: each sketched person's position, screen size, pose and the gaze/motion arrows. Ignore the sketch's background lines. The CHARACTER REFERENCE photographs show the real people.
Place the real people into the background at exactly the sketched positions, sizes and poses, following the arrow directions. No sketch lines or arrows may remain.

Create ONE FINAL photorealistic live-action film still of the moment below — contemporary South Korea, 2026; all people are Korean unless stated. TIME OF DAY (lock): early morning.

SHOT TEXT (authoritative, Korean): 책상 앞 모니터를 향해 앉은 40대 한국인 남성 보안직원과 혜수의 뒷모습, 그 뒤에 선 수리영의 굳은 정면 구도

LOCATION (lock): Inside the supermarket's security office, at a desk with a CCTV monitor under office lighting. The shot takes place here — the attached LOCATION PHOTOGRAPH shows the exact spot.

CAMERA & FRAME (follow exactly — this is the composition authority for this still; the reference images supply identity and place, never the framing):
- CAMERA: Hold a static medium-wide composition from seated eye level behind the security employee and Hyesu, placing their backs across the lower left and right while the monitor sits between their shoulder lines. Suriyoung remains rigidly frontal in the central background, creating a compressed line of monitor, seated observers, and anxious witness before the later forward move.
- FRAMING SCALE: wide shot
- FRAME LAYOUT: 40대 한국인 남성 보안직원 in the lower-left of the frame, foreground; 혜수 in the lower-right of the frame, foreground; 수리영 in the upper-center of the frame, background, looks toward CCTV monitor.
- KEY BACKGROUND ELEMENTS: CCTV 모니터 (On and replaying footage) — Its active front face is visible, displaying CCTV playback with one section markedly darker and difficult to identify; used as Central visual anchor framed between the two seated figures; 책상 (In use) — The monitor-facing edge runs across the frame between the seated figures and Suriyoung; used as Supports the monitor and establishes the office work area across the midground; 의자 두 개 (Occupied) — The backs face the camera while the seats face the monitor; used as Forms two foreground blocks that compress Suriyoung in depth.
Compose the frame exactly as specified above — camera angle, subject scale and screen placement. Keep true physical scale between people and background elements: fixtures and distant objects occupy only the small screen area their real size and distance dictate; never enlarge a background object into a foreground presence.

REALIZE FIGURATIVE LANGUAGE AS A LIVE-ACTION SHOT: the Korean shot
text may describe characters metaphorically, figuratively or with
exaggeration. Photograph what a real movie camera would actually
record on a physical set — exaggerated or figurative impressions
become realistic staging choices (lighting, distance, angle,
wardrobe), not literal fantasy imagery.
EVERY CHARACTER IS A HUMAN BEING: unless the story explicitly
features non-human or virtual beings (as in science fiction or
fantasy), every character — however indirectly, vaguely or
figuratively the text describes them — IS a real human. When the
text gives no direct visual description of a person, IMAGINE one
and still show them as a concrete, fully-formed human being:
always render the human form to the maximum extent — build,
posture, face, hands, clothing — never reduce a person to a shape,
blob, solid silhouette or abstract mass.

EXPRESSIONS ARE ACTED, NEVER ANATOMICAL: when the text describes a
person's eyes, face or presence emotionally or figuratively (vacant,
hollow, dazed, burning, lifeless gaze and the like), it describes an
ACTOR'S PERFORMANCE captured by a real camera — realize it ONLY
through gaze direction, focus, eyelids, facial muscles, stillness
and posture. Every living person's eyes remain anatomically normal
human eyes with a natural iris and pupil, natural sclera and normal
proportions; NEVER whiten, blank out, cloud over, glow, enlarge or
otherwise alter eyeballs, skin or anatomy — unless the story
explicitly declares that being non-human or supernatural in form.

PROPS FACE THE RIGHT WAY: every handheld or used object must be
oriented exactly as its real-world use requires. A person reading,
watching or operating something (a phone, a photograph, a paper,
any device) has its functional side — screen, front, page — facing
THEIR OWN eyes; the camera then sees whatever side the staging
geometry implies (often its back). Show the functional side to the
camera ONLY when the shot text itself stages it toward the viewer.
Never flip, mirror or reverse an object's front and back.

CARRIED STATE (persist exactly — must match the neighbouring shots of this scene): The relevant section of the mart CCTV recording remains uniquely dark and impossible to identify through repeated playback. Su-ri-young still has her doll-adorned backpack with her.

PEOPLE: the SHOT TEXT alone decides whether any person is visible in this shot. IF a person appears, they must be one of: 혜수 (한국인 40대 초반 여성, 단정한 검은 머리, 성숙한 타원형 얼굴); 수리영 (한국인 20대 초반 여성, 긴 검은 생머리, 젊은 타원형 얼굴); 40대 한국인 남성 보안직원 (한국인 40대 남성, 짧은 검은 머리, 단정한 머리 모양, 중년 남성 얼굴) — never anyone else, and never add a person the shot text does not show. IF only part of a person is in frame (a hand, arm, foot, back, silhouette), that body part belongs to the specific person the shot text names — its sex, age, build, skin and grooming must unmistakably match that person's profile above.

No text, captions, watermarks or annotations anywhere.

Create ONE FINAL photorealistic live-action film still of the moment below — contemporary South Korea, 2026; all people are Korean unless stated. TIME OF DAY (lock): early morning.

SHOT TEXT (authoritative, Korean): 책상 앞 모니터를 향해 앉은 40대 한국인 남성 보안직원과 혜수의 뒷모습, 그 뒤에 선 수리영의 굳은 정면 구도

LOCATION (lock): Inside the supermarket's security office, at a desk with a CCTV monitor under office lighting. The shot takes place here — the attached LOCATION PHOTOGRAPH shows the exact spot.

CAMERA & FRAME (follow exactly — this is the composition authority for this still; the reference images supply identity and place, never the framing):
- CAMERA: Hold a static medium-wide composition from seated eye level behind the security employee and Hyesu, placing their backs across the lower left and right while the monitor sits between their shoulder lines. Suriyoung remains rigidly frontal in the central background, creating a compressed line of monitor, seated observers, and anxious witness before the later forward move.
- FRAMING SCALE: wide shot
- FRAME LAYOUT: 40대 한국인 남성 보안직원 in the lower-left of the frame, foreground; 혜수 in the lower-right of the frame, foreground; 수리영 in the upper-center of the frame, background, looks toward CCTV monitor.
- KEY BACKGROUND ELEMENTS: CCTV 모니터 (On and replaying footage) — Its active front face is visible, displaying CCTV playback with one section markedly darker and difficult to identify; used as Central visual anchor framed between the two seated figures; 책상 (In use) — The monitor-facing edge runs across the frame between the seated figures and Suriyoung; used as Supports the monitor and establishes the office work area across the midground; 의자 두 개 (Occupied) — The backs face the camera while the seats face the monitor; used as Forms two foreground blocks that compress Suriyoung in depth.
Compose the frame exactly as specified above — camera angle, subject scale and screen placement. Keep true physical scale between people and background elements: fixtures and distant objects occupy only the small screen area their real size and distance dictate; never enlarge a background object into a foreground presence.

REALIZE FIGURATIVE LANGUAGE AS A LIVE-ACTION SHOT: the Korean shot
text may describe characters metaphorically, figuratively or with
exaggeration. Photograph what a real movie camera would actually
record on a physical set — exaggerated or figurative impressions
become realistic staging choices (lighting, distance, angle,
wardrobe), not literal fantasy imagery.
EVERY CHARACTER IS A HUMAN BEING: unless the story explicitly
features non-human or virtual beings (as in science fiction or
fantasy), every character — however indirectly, vaguely or
figuratively the text describes them — IS a real human. When the
text gives no direct visual description of a person, IMAGINE one
and still show them as a concrete, fully-formed human being:
always render the human form to the maximum extent — build,
posture, face, hands, clothing — never reduce a person to a shape,
blob, solid silhouette or abstract mass.

EXPRESSIONS ARE ACTED, NEVER ANATOMICAL: when the text describes a
person's eyes, face or presence emotionally or figuratively (vacant,
hollow, dazed, burning, lifeless gaze and the like), it describes an
ACTOR'S PERFORMANCE captured by a real camera — realize it ONLY
through gaze direction, focus, eyelids, facial muscles, stillness
and posture. Every living person's eyes remain anatomically normal
human eyes with a natural iris and pupil, natural sclera and normal
proportions; NEVER whiten, blank out, cloud over, glow, enlarge or
otherwise alter eyeballs, skin or anatomy — unless the story
explicitly declares that being non-human or supernatural in form.

PROPS FACE THE RIGHT WAY: every handheld or used object must be
oriented exactly as its real-world use requires. A person reading,
watching or operating something (a phone, a photograph, a paper,
any device) has its functional side — screen, front, page — facing
THEIR OWN eyes; the camera then sees whatever side the staging
geometry implies (often its back). Show the functional side to the
camera ONLY when the shot text itself stages it toward the viewer.
Never flip, mirror or reverse an object's front and back.

CARRIED STATE (persist exactly — must match the neighbouring shots of this scene): The relevant section of the mart CCTV recording remains uniquely dark and impossible to identify through repeated playback. Su-ri-young still has her doll-adorned backpack with her.

PEOPLE: the SHOT TEXT alone decides whether any person is visible in this shot. IF a person appears, they must be one of: 혜수 (한국인 40대 초반 여성, 단정한 검은 머리, 성숙한 타원형 얼굴); 수리영 (한국인 20대 초반 여성, 긴 검은 생머리, 젊은 타원형 얼굴); 40대 한국인 남성 보안직원 (한국인 40대 남성, 짧은 검은 머리, 단정한 머리 모양, 중년 남성 얼굴) — never anyone else, and never add a person the shot text does not show. IF only part of a person is in frame (a hand, arm, foot, back, silhouette), that body part belongs to the specific person the shot text names — its sex, age, build, skin and grooming must unmistakably match that person's profile above.

No text, captions, watermarks or annotations anywhere.

Edit this photograph. Apply ONLY the corrections listed below,
each in place, changing nothing else:

CORRECTIONS:
- Redirect Suriyoung's eyes downwards so she is looking directly at the CCTV monitor on the desk.

PRESERVE EVERYTHING ELSE EXACTLY as in the original photograph —
every person, pose, face, garment, object, set feature, lighting,
camera and framing stay identical. No text, captions or
watermarks anywhere.

Create ONE FINAL photorealistic live-action film still of the moment below — contemporary South Korea, 2026; all people are Korean unless stated. TIME OF DAY (lock): dusk.

SHOT TEXT (authoritative, Korean): 마트 밖으로 나온 수리영이 휴대폰을 귀에 댄 채 불안한 표정으로 멈춰 선 전신

LOCATION (lock): Just outside the large supermarket entrance at dusk, where she stops to make a call. The shot takes place here — the attached LOCATION PHOTOGRAPH shows the exact spot.

CAMERA & FRAME (follow exactly — this is the composition authority for this still; the reference images supply identity and place, never the framing):
- CAMERA: Outside the store, the camera tracks several meters ahead and slightly to the side of 수리영 before settling into a locked full-body hold when she stops. She occupies the middle-right of frame with the phone pressed to her ear, while the store exterior and commercial street retain enough depth around her to make the sudden halt feel exposed.
- FRAMING SCALE: wide shot
- KEY BACKGROUND ELEMENTS: 대형 마트 외부 (Exterior visible; entrance state not specified) — The entrance-facing side sits behind and to the left of 수리영; used as Background spatial anchor showing that she has just exited; 번화가 네온사인 (Present along the commercial street) — Several sign faces are angled toward the camera; their specific text and operating state are not established; used as Layered background activity around the isolated full-body figure; 택시 (Present on the street) — Their side and three-quarter faces pass through the deeper street plane; used as Moving background context beyond the stopped subject.
Compose the frame exactly as specified above — camera angle, subject scale and screen placement. Keep true physical scale between people and background elements: fixtures and distant objects occupy only the small screen area their real size and distance dictate; never enlarge a background object into a foreground presence.

REALIZE FIGURATIVE LANGUAGE AS A LIVE-ACTION SHOT: the Korean shot
text may describe characters metaphorically, figuratively or with
exaggeration. Photograph what a real movie camera would actually
record on a physical set — exaggerated or figurative impressions
become realistic staging choices (lighting, distance, angle,
wardrobe), not literal fantasy imagery.
EVERY CHARACTER IS A HUMAN BEING: unless the story explicitly
features non-human or virtual beings (as in science fiction or
fantasy), every character — however indirectly, vaguely or
figuratively the text describes them — IS a real human. When the
text gives no direct visual description of a person, IMAGINE one
and still show them as a concrete, fully-formed human being:
always render the human form to the maximum extent — build,
posture, face, hands, clothing — never reduce a person to a shape,
blob, solid silhouette or abstract mass.

EXPRESSIONS ARE ACTED, NEVER ANATOMICAL: when the text describes a
person's eyes, face or presence emotionally or figuratively (vacant,
hollow, dazed, burning, lifeless gaze and the like), it describes an
ACTOR'S PERFORMANCE captured by a real camera — realize it ONLY
through gaze direction, focus, eyelids, facial muscles, stillness
and posture. Every living person's eyes remain anatomically normal
human eyes with a natural iris and pupil, natural sclera and normal
proportions; NEVER whiten, blank out, cloud over, glow, enlarge or
otherwise alter eyeballs, skin or anatomy — unless the story
explicitly declares that being non-human or supernatural in form.

PROPS FACE THE RIGHT WAY: every handheld or used object must be
oriented exactly as its real-world use requires. A person reading,
watching or operating something (a phone, a photograph, a paper,
any device) has its functional side — screen, front, page — facing
THEIR OWN eyes; the camera then sees whatever side the staging
geometry implies (often its back). Show the functional side to the
camera ONLY when the shot text itself stages it toward the viewer.
Never flip, mirror or reverse an object's front and back.

CARRIED STATE (persist exactly — must match the neighbouring shots of this scene): Su-ri-young still has her phone and is beside the bicycle she parked at the mart, ready to head home.

PEOPLE: the SHOT TEXT alone decides whether any person is visible in this shot. IF a person appears, they must be one of: 수리영 (한국인 20대 초반 여성, 긴 검은 생머리, 젊은 타원형 얼굴) — never anyone else, and never add a person the shot text does not show. IF only part of a person is in frame (a hand, arm, foot, back, silhouette), that body part belongs to the specific person the shot text names — its sex, age, build, skin and grooming must unmistakably match that person's profile above.

No text, captions, watermarks or annotations anywhere.

Stage the shot. The FIRST attached image (SHOT BACKGROUND) is the finished empty background of this shot — keep it EXACTLY: its camera, perspective, architecture, lighting and every fixed feature stay untouched. The SECOND attached image (LAYOUT SKETCH) tells you ONLY where the people go: each sketched person's position, screen size, pose and the gaze/motion arrows. Ignore the sketch's background lines. The CHARACTER REFERENCE photographs show the real people.
Place the real people into the background at exactly the sketched positions, sizes and poses, following the arrow directions. No sketch lines or arrows may remain.

Create ONE FINAL photorealistic live-action film still of the moment below — contemporary South Korea, 2026; all people are Korean unless stated. TIME OF DAY (lock): dusk.

SHOT TEXT (authoritative, Korean): 마트 밖으로 나온 수리영이 휴대폰을 귀에 댄 채 불안한 표정으로 멈춰 선 전신

LOCATION (lock): Just outside the large supermarket entrance at dusk, where she stops to make a call. The shot takes place here — the attached LOCATION PHOTOGRAPH shows the exact spot.

CAMERA & FRAME (follow exactly — this is the composition authority for this still; the reference images supply identity and place, never the framing):
- CAMERA: Outside the store, the camera tracks several meters ahead and slightly to the side of 수리영 before settling into a locked full-body hold when she stops. She occupies the middle-right of frame with the phone pressed to her ear, while the store exterior and commercial street retain enough depth around her to make the sudden halt feel exposed.
- FRAMING SCALE: wide shot
- KEY BACKGROUND ELEMENTS: 대형 마트 외부 (Exterior visible; entrance state not specified) — The entrance-facing side sits behind and to the left of 수리영; used as Background spatial anchor showing that she has just exited; 번화가 네온사인 (Present along the commercial street) — Several sign faces are angled toward the camera; their specific text and operating state are not established; used as Layered background activity around the isolated full-body figure; 택시 (Present on the street) — Their side and three-quarter faces pass through the deeper street plane; used as Moving background context beyond the stopped subject.
Compose the frame exactly as specified above — camera angle, subject scale and screen placement. Keep true physical scale between people and background elements: fixtures and distant objects occupy only the small screen area their real size and distance dictate; never enlarge a background object into a foreground presence.

REALIZE FIGURATIVE LANGUAGE AS A LIVE-ACTION SHOT: the Korean shot
text may describe characters metaphorically, figuratively or with
exaggeration. Photograph what a real movie camera would actually
record on a physical set — exaggerated or figurative impressions
become realistic staging choices (lighting, distance, angle,
wardrobe), not literal fantasy imagery.
EVERY CHARACTER IS A HUMAN BEING: unless the story explicitly
features non-human or virtual beings (as in science fiction or
fantasy), every character — however indirectly, vaguely or
figuratively the text describes them — IS a real human. When the
text gives no direct visual description of a person, IMAGINE one
and still show them as a concrete, fully-formed human being:
always render the human form to the maximum extent — build,
posture, face, hands, clothing — never reduce a person to a shape,
blob, solid silhouette or abstract mass.

EXPRESSIONS ARE ACTED, NEVER ANATOMICAL: when the text describes a
person's eyes, face or presence emotionally or figuratively (vacant,
hollow, dazed, burning, lifeless gaze and the like), it describes an
ACTOR'S PERFORMANCE captured by a real camera — realize it ONLY
through gaze direction, focus, eyelids, facial muscles, stillness
and posture. Every living person's eyes remain anatomically normal
human eyes with a natural iris and pupil, natural sclera and normal
proportions; NEVER whiten, blank out, cloud over, glow, enlarge or
otherwise alter eyeballs, skin or anatomy — unless the story
explicitly declares that being non-human or supernatural in form.

PROPS FACE THE RIGHT WAY: every handheld or used object must be
oriented exactly as its real-world use requires. A person reading,
watching or operating something (a phone, a photograph, a paper,
any device) has its functional side — screen, front, page — facing
THEIR OWN eyes; the camera then sees whatever side the staging
geometry implies (often its back). Show the functional side to the
camera ONLY when the shot text itself stages it toward the viewer.
Never flip, mirror or reverse an object's front and back.

CARRIED STATE (persist exactly — must match the neighbouring shots of this scene): Su-ri-young still has her phone and is beside the bicycle she parked at the mart, ready to head home.

PEOPLE: the SHOT TEXT alone decides whether any person is visible in this shot. IF a person appears, they must be one of: 수리영 (한국인 20대 초반 여성, 긴 검은 생머리, 젊은 타원형 얼굴) — never anyone else, and never add a person the shot text does not show. IF only part of a person is in frame (a hand, arm, foot, back, silhouette), that body part belongs to the specific person the shot text names — its sex, age, build, skin and grooming must unmistakably match that person's profile above.

No text, captions, watermarks or annotations anywhere.

Create ONE FINAL photorealistic live-action film still of the moment below — contemporary South Korea, 2026; all people are Korean unless stated. TIME OF DAY (lock): dusk.

SHOT TEXT (authoritative, Korean): 마트 밖으로 나온 수리영이 휴대폰을 귀에 댄 채 불안한 표정으로 멈춰 선 전신

LOCATION (lock): Just outside the large supermarket entrance at dusk, where she stops to make a call. The shot takes place here — the attached LOCATION PHOTOGRAPH shows the exact spot.

CAMERA & FRAME (follow exactly — this is the composition authority for this still; the reference images supply identity and place, never the framing):
- CAMERA: Outside the store, the camera tracks several meters ahead and slightly to the side of 수리영 before settling into a locked full-body hold when she stops. She occupies the middle-right of frame with the phone pressed to her ear, while the store exterior and commercial street retain enough depth around her to make the sudden halt feel exposed.
- FRAMING SCALE: wide shot
- KEY BACKGROUND ELEMENTS: 대형 마트 외부 (Exterior visible; entrance state not specified) — The entrance-facing side sits behind and to the left of 수리영; used as Background spatial anchor showing that she has just exited; 번화가 네온사인 (Present along the commercial street) — Several sign faces are angled toward the camera; their specific text and operating state are not established; used as Layered background activity around the isolated full-body figure; 택시 (Present on the street) — Their side and three-quarter faces pass through the deeper street plane; used as Moving background context beyond the stopped subject.
Compose the frame exactly as specified above — camera angle, subject scale and screen placement. Keep true physical scale between people and background elements: fixtures and distant objects occupy only the small screen area their real size and distance dictate; never enlarge a background object into a foreground presence.

REALIZE FIGURATIVE LANGUAGE AS A LIVE-ACTION SHOT: the Korean shot
text may describe characters metaphorically, figuratively or with
exaggeration. Photograph what a real movie camera would actually
record on a physical set — exaggerated or figurative impressions
become realistic staging choices (lighting, distance, angle,
wardrobe), not literal fantasy imagery.
EVERY CHARACTER IS A HUMAN BEING: unless the story explicitly
features non-human or virtual beings (as in science fiction or
fantasy), every character — however indirectly, vaguely or
figuratively the text describes them — IS a real human. When the
text gives no direct visual description of a person, IMAGINE one
and still show them as a concrete, fully-formed human being:
always render the human form to the maximum extent — build,
posture, face, hands, clothing — never reduce a person to a shape,
blob, solid silhouette or abstract mass.

EXPRESSIONS ARE ACTED, NEVER ANATOMICAL: when the text describes a
person's eyes, face or presence emotionally or figuratively (vacant,
hollow, dazed, burning, lifeless gaze and the like), it describes an
ACTOR'S PERFORMANCE captured by a real camera — realize it ONLY
through gaze direction, focus, eyelids, facial muscles, stillness
and posture. Every living person's eyes remain anatomically normal
human eyes with a natural iris and pupil, natural sclera and normal
proportions; NEVER whiten, blank out, cloud over, glow, enlarge or
otherwise alter eyeballs, skin or anatomy — unless the story
explicitly declares that being non-human or supernatural in form.

PROPS FACE THE RIGHT WAY: every handheld or used object must be
oriented exactly as its real-world use requires. A person reading,
watching or operating something (a phone, a photograph, a paper,
any device) has its functional side — screen, front, page — facing
THEIR OWN eyes; the camera then sees whatever side the staging
geometry implies (often its back). Show the functional side to the
camera ONLY when the shot text itself stages it toward the viewer.
Never flip, mirror or reverse an object's front and back.

CARRIED STATE (persist exactly — must match the neighbouring shots of this scene): Su-ri-young still has her phone and is beside the bicycle she parked at the mart, ready to head home.

PEOPLE: the SHOT TEXT alone decides whether any person is visible in this shot. IF a person appears, they must be one of: 수리영 (한국인 20대 초반 여성, 긴 검은 생머리, 젊은 타원형 얼굴) — never anyone else, and never add a person the shot text does not show. IF only part of a person is in frame (a hand, arm, foot, back, silhouette), that body part belongs to the specific person the shot text names — its sex, age, build, skin and grooming must unmistakably match that person's profile above.

No text, captions, watermarks or annotations anywhere.

Edit this photograph. Apply ONLY the corrections listed below,
each in place, changing nothing else:

CORRECTIONS:
- Add the green cap from the character reference image to the subject's head.
- Remove all text from the neon signs and banners along the street, leaving them blank.

PRESERVE EVERYTHING ELSE EXACTLY as in the original photograph —
every person, pose, face, garment, object, set feature, lighting,
camera and framing stay identical. No text, captions or
watermarks anywhere.

Create ONE FINAL photorealistic live-action film still of the moment below — contemporary South Korea, 2026; all people are Korean unless stated. TIME OF DAY (lock): night.

SHOT TEXT (authoritative, Korean): 커튼이 젖혀진 바닥에 강민숙이 고개를 떨구고 앉은 채 미동도 하지 않는 정지 상태

LOCATION (lock): Inside the daughter's bedroom, on the bloodied floor behind the opened curtain. Only faint spill from the living-room floor lamp reaches the nighttime apartment. The shot takes place here — the attached LOCATION PHOTOGRAPH shows the exact spot.

CAMERA & FRAME (follow exactly — this is the composition authority for this still; the reference images supply identity and place, never the framing):
- CAMERA: At knee height facing the opened curtain, the static camera frames 강민숙’s seated form from the lowered head through the collapsed legs, with the curtain edges forming narrow foreground brackets. She occupies the central midground while visible blood on the floor and her injured shoulder supply the only sharp crimson accents within the otherwise restrained room.
- FRAMING SCALE: medium shot
- KEY BACKGROUND ELEMENTS: 커튼 (Pulled aside) — Its drawn-aside edges face inward toward the revealed seated figure; used as Opened foreground frame that reveals the body; 바닥의 혈흔 (Pooled across the floor with bloody footprints leading to the curtain); used as Lower-frame path leading visually to 강민숙.
Compose the frame exactly as specified above — camera angle, subject scale and screen placement. Keep true physical scale between people and background elements: fixtures and distant objects occupy only the small screen area their real size and distance dictate; never enlarge a background object into a foreground presence.

IMMOBILE CHARACTER POSE — CANONICAL (identical wherever this character appears in ANY panel; on any conflict THIS POSE WINS): She is seated motionless on the floor behind the curtain with her torso slumped and her head dropped forward; one rigid hand is closed around a crumpled photograph.

IMMOBILE BODIES OBEY GRAVITY: a person who is dead or unconscious
exerts NO muscular effort. Every part of their body — head, torso,
arms, hands, fingers, legs — rests fully on whatever supports it
(floor, wall, furniture, their own lap) and hangs or slumps with
gravity. NEVER show any part of an immobile person's body lifted,
raised, held up in the air, or posed as if presenting something:
an object in their grip stays clenched in a hand that itself lies
fallen on a support — the hand does not hold the object up. If the
canonical pose leaves a body part unspecified, resolve it as the
most gravity-compliant, fully-supported position.

REALIZE FIGURATIVE LANGUAGE AS A LIVE-ACTION SHOT: the Korean shot
text may describe characters metaphorically, figuratively or with
exaggeration. Photograph what a real movie camera would actually
record on a physical set — exaggerated or figurative impressions
become realistic staging choices (lighting, distance, angle,
wardrobe), not literal fantasy imagery.
EVERY CHARACTER IS A HUMAN BEING: unless the story explicitly
features non-human or virtual beings (as in science fiction or
fantasy), every character — however indirectly, vaguely or
figuratively the text describes them — IS a real human. When the
text gives no direct visual description of a person, IMAGINE one
and still show them as a concrete, fully-formed human being:
always render the human form to the maximum extent — build,
posture, face, hands, clothing — never reduce a person to a shape,
blob, solid silhouette or abstract mass.

EXPRESSIONS ARE ACTED, NEVER ANATOMICAL: when the text describes a
person's eyes, face or presence emotionally or figuratively (vacant,
hollow, dazed, burning, lifeless gaze and the like), it describes an
ACTOR'S PERFORMANCE captured by a real camera — realize it ONLY
through gaze direction, focus, eyelids, facial muscles, stillness
and posture. Every living person's eyes remain anatomically normal
human eyes with a natural iris and pupil, natural sclera and normal
proportions; NEVER whiten, blank out, cloud over, glow, enlarge or
otherwise alter eyeballs, skin or anatomy — unless the story
explicitly declares that being non-human or supernatural in form.

PROPS FACE THE RIGHT WAY: every handheld or used object must be
oriented exactly as its real-world use requires. A person reading,
watching or operating something (a phone, a photograph, a paper,
any device) has its functional side — screen, front, page — facing
THEIR OWN eyes; the camera then sees whatever side the staging
geometry implies (often its back). Show the functional side to the
camera ONLY when the shot text itself stages it toward the viewer.
Never flip, mirror or reverse an object's front and back.

CARRIED STATE (persist exactly — must match the neighbouring shots of this scene): Min-suk is seated motionless on the floor behind the opened curtain with her torso slumped and her head dropped forward, amid the blood and bloody footprints. One rigid hand is closed around the crumpled photograph later revealed in close-up, and the circular mark remains on her wrist.

PEOPLE: the SHOT TEXT alone decides whether any person is visible in this shot. IF a person appears, they must be one of: 강민숙 (한국인 40대 초반 여성, 단정한 검은 머리, 자연스러운 중년 얼굴) — never anyone else, and never add a person the shot text does not show. IF only part of a person is in frame (a hand, arm, foot, back, silhouette), that body part belongs to the specific person the shot text names — its sex, age, build, skin and grooming must unmistakably match that person's profile above.

No text, captions, watermarks or annotations anywhere.

Stage the shot. The FIRST attached image (SHOT BACKGROUND) is the finished empty background of this shot — keep it EXACTLY: its camera, perspective, architecture, lighting and every fixed feature stay untouched. The SECOND attached image (LAYOUT SKETCH) tells you ONLY where the people go: each sketched person's position, screen size, pose and the gaze/motion arrows. Ignore the sketch's background lines. The CHARACTER REFERENCE photographs show the real people.
Place the real people into the background at exactly the sketched positions, sizes and poses, following the arrow directions. No sketch lines or arrows may remain.

Create ONE FINAL photorealistic live-action film still of the moment below — contemporary South Korea, 2026; all people are Korean unless stated. TIME OF DAY (lock): night.

SHOT TEXT (authoritative, Korean): 커튼이 젖혀진 바닥에 강민숙이 고개를 떨구고 앉은 채 미동도 하지 않는 정지 상태

LOCATION (lock): Inside the daughter's bedroom, on the bloodied floor behind the opened curtain. Only faint spill from the living-room floor lamp reaches the nighttime apartment. The shot takes place here — the attached LOCATION PHOTOGRAPH shows the exact spot.

CAMERA & FRAME (follow exactly — this is the composition authority for this still; the reference images supply identity and place, never the framing):
- CAMERA: At knee height facing the opened curtain, the static camera frames 강민숙’s seated form from the lowered head through the collapsed legs, with the curtain edges forming narrow foreground brackets. She occupies the central midground while visible blood on the floor and her injured shoulder supply the only sharp crimson accents within the otherwise restrained room.
- FRAMING SCALE: medium shot
- KEY BACKGROUND ELEMENTS: 커튼 (Pulled aside) — Its drawn-aside edges face inward toward the revealed seated figure; used as Opened foreground frame that reveals the body; 바닥의 혈흔 (Pooled across the floor with bloody footprints leading to the curtain); used as Lower-frame path leading visually to 강민숙.
Compose the frame exactly as specified above — camera angle, subject scale and screen placement. Keep true physical scale between people and background elements: fixtures and distant objects occupy only the small screen area their real size and distance dictate; never enlarge a background object into a foreground presence.

IMMOBILE CHARACTER POSE — CANONICAL (identical wherever this character appears in ANY panel; on any conflict THIS POSE WINS): She is seated motionless on the floor behind the curtain with her torso slumped and her head dropped forward; one rigid hand is closed around a crumpled photograph.

IMMOBILE BODIES OBEY GRAVITY: a person who is dead or unconscious
exerts NO muscular effort. Every part of their body — head, torso,
arms, hands, fingers, legs — rests fully on whatever supports it
(floor, wall, furniture, their own lap) and hangs or slumps with
gravity. NEVER show any part of an immobile person's body lifted,
raised, held up in the air, or posed as if presenting something:
an object in their grip stays clenched in a hand that itself lies
fallen on a support — the hand does not hold the object up. If the
canonical pose leaves a body part unspecified, resolve it as the
most gravity-compliant, fully-supported position.

REALIZE FIGURATIVE LANGUAGE AS A LIVE-ACTION SHOT: the Korean shot
text may describe characters metaphorically, figuratively or with
exaggeration. Photograph what a real movie camera would actually
record on a physical set — exaggerated or figurative impressions
become realistic staging choices (lighting, distance, angle,
wardrobe), not literal fantasy imagery.
EVERY CHARACTER IS A HUMAN BEING: unless the story explicitly
features non-human or virtual beings (as in science fiction or
fantasy), every character — however indirectly, vaguely or
figuratively the text describes them — IS a real human. When the
text gives no direct visual description of a person, IMAGINE one
and still show them as a concrete, fully-formed human being:
always render the human form to the maximum extent — build,
posture, face, hands, clothing — never reduce a person to a shape,
blob, solid silhouette or abstract mass.

EXPRESSIONS ARE ACTED, NEVER ANATOMICAL: when the text describes a
person's eyes, face or presence emotionally or figuratively (vacant,
hollow, dazed, burning, lifeless gaze and the like), it describes an
ACTOR'S PERFORMANCE captured by a real camera — realize it ONLY
through gaze direction, focus, eyelids, facial muscles, stillness
and posture. Every living person's eyes remain anatomically normal
human eyes with a natural iris and pupil, natural sclera and normal
proportions; NEVER whiten, blank out, cloud over, glow, enlarge or
otherwise alter eyeballs, skin or anatomy — unless the story
explicitly declares that being non-human or supernatural in form.

PROPS FACE THE RIGHT WAY: every handheld or used object must be
oriented exactly as its real-world use requires. A person reading,
watching or operating something (a phone, a photograph, a paper,
any device) has its functional side — screen, front, page — facing
THEIR OWN eyes; the camera then sees whatever side the staging
geometry implies (often its back). Show the functional side to the
camera ONLY when the shot text itself stages it toward the viewer.
Never flip, mirror or reverse an object's front and back.

CARRIED STATE (persist exactly — must match the neighbouring shots of this scene): Min-suk is seated motionless on the floor behind the opened curtain with her torso slumped and her head dropped forward, amid the blood and bloody footprints. One rigid hand is closed around the crumpled photograph later revealed in close-up, and the circular mark remains on her wrist.

PEOPLE: the SHOT TEXT alone decides whether any person is visible in this shot. IF a person appears, they must be one of: 강민숙 (한국인 40대 초반 여성, 단정한 검은 머리, 자연스러운 중년 얼굴) — never anyone else, and never add a person the shot text does not show. IF only part of a person is in frame (a hand, arm, foot, back, silhouette), that body part belongs to the specific person the shot text names — its sex, age, build, skin and grooming must unmistakably match that person's profile above.

No text, captions, watermarks or annotations anywhere.

Create ONE FINAL photorealistic live-action film still of the moment below — contemporary South Korea, 2026; all people are Korean unless stated. TIME OF DAY (lock): night.

SHOT TEXT (authoritative, Korean): 커튼이 젖혀진 바닥에 강민숙이 고개를 떨구고 앉은 채 미동도 하지 않는 정지 상태

LOCATION (lock): Inside the daughter's bedroom, on the bloodied floor behind the opened curtain. Only faint spill from the living-room floor lamp reaches the nighttime apartment. The shot takes place here — the attached LOCATION PHOTOGRAPH shows the exact spot.

CAMERA & FRAME (follow exactly — this is the composition authority for this still; the reference images supply identity and place, never the framing):
- CAMERA: At knee height facing the opened curtain, the static camera frames 강민숙’s seated form from the lowered head through the collapsed legs, with the curtain edges forming narrow foreground brackets. She occupies the central midground while visible blood on the floor and her injured shoulder supply the only sharp crimson accents within the otherwise restrained room.
- FRAMING SCALE: medium shot
- KEY BACKGROUND ELEMENTS: 커튼 (Pulled aside) — Its drawn-aside edges face inward toward the revealed seated figure; used as Opened foreground frame that reveals the body; 바닥의 혈흔 (Pooled across the floor with bloody footprints leading to the curtain); used as Lower-frame path leading visually to 강민숙.
Compose the frame exactly as specified above — camera angle, subject scale and screen placement. Keep true physical scale between people and background elements: fixtures and distant objects occupy only the small screen area their real size and distance dictate; never enlarge a background object into a foreground presence.

IMMOBILE CHARACTER POSE — CANONICAL (identical wherever this character appears in ANY panel; on any conflict THIS POSE WINS): She is seated motionless on the floor behind the curtain with her torso slumped and her head dropped forward; one rigid hand is closed around a crumpled photograph.

IMMOBILE BODIES OBEY GRAVITY: a person who is dead or unconscious
exerts NO muscular effort. Every part of their body — head, torso,
arms, hands, fingers, legs — rests fully on whatever supports it
(floor, wall, furniture, their own lap) and hangs or slumps with
gravity. NEVER show any part of an immobile person's body lifted,
raised, held up in the air, or posed as if presenting something:
an object in their grip stays clenched in a hand that itself lies
fallen on a support — the hand does not hold the object up. If the
canonical pose leaves a body part unspecified, resolve it as the
most gravity-compliant, fully-supported position.

REALIZE FIGURATIVE LANGUAGE AS A LIVE-ACTION SHOT: the Korean shot
text may describe characters metaphorically, figuratively or with
exaggeration. Photograph what a real movie camera would actually
record on a physical set — exaggerated or figurative impressions
become realistic staging choices (lighting, distance, angle,
wardrobe), not literal fantasy imagery.
EVERY CHARACTER IS A HUMAN BEING: unless the story explicitly
features non-human or virtual beings (as in science fiction or
fantasy), every character — however indirectly, vaguely or
figuratively the text describes them — IS a real human. When the
text gives no direct visual description of a person, IMAGINE one
and still show them as a concrete, fully-formed human being:
always render the human form to the maximum extent — build,
posture, face, hands, clothing — never reduce a person to a shape,
blob, solid silhouette or abstract mass.

EXPRESSIONS ARE ACTED, NEVER ANATOMICAL: when the text describes a
person's eyes, face or presence emotionally or figuratively (vacant,
hollow, dazed, burning, lifeless gaze and the like), it describes an
ACTOR'S PERFORMANCE captured by a real camera — realize it ONLY
through gaze direction, focus, eyelids, facial muscles, stillness
and posture. Every living person's eyes remain anatomically normal
human eyes with a natural iris and pupil, natural sclera and normal
proportions; NEVER whiten, blank out, cloud over, glow, enlarge or
otherwise alter eyeballs, skin or anatomy — unless the story
explicitly declares that being non-human or supernatural in form.

PROPS FACE THE RIGHT WAY: every handheld or used object must be
oriented exactly as its real-world use requires. A person reading,
watching or operating something (a phone, a photograph, a paper,
any device) has its functional side — screen, front, page — facing
THEIR OWN eyes; the camera then sees whatever side the staging
geometry implies (often its back). Show the functional side to the
camera ONLY when the shot text itself stages it toward the viewer.
Never flip, mirror or reverse an object's front and back.

CARRIED STATE (persist exactly — must match the neighbouring shots of this scene): Min-suk is seated motionless on the floor behind the opened curtain with her torso slumped and her head dropped forward, amid the blood and bloody footprints. One rigid hand is closed around the crumpled photograph later revealed in close-up, and the circular mark remains on her wrist.

PEOPLE: the SHOT TEXT alone decides whether any person is visible in this shot. IF a person appears, they must be one of: 강민숙 (한국인 40대 초반 여성, 단정한 검은 머리, 자연스러운 중년 얼굴) — never anyone else, and never add a person the shot text does not show. IF only part of a person is in frame (a hand, arm, foot, back, silhouette), that body part belongs to the specific person the shot text names — its sex, age, build, skin and grooming must unmistakably match that person's profile above.

No text, captions, watermarks or annotations anywhere.

Edit this photograph. Apply ONLY the corrections listed below,
each in place, changing nothing else:

CORRECTIONS:
- Crumple the photograph into a tight ball and change the hand into a lifeless, closed fist completely wrapped around the paper, resting limply on her leg.

PRESERVE EVERYTHING ELSE EXACTLY as in the original photograph —
every person, pose, face, garment, object, set feature, lighting,
camera and framing stay identical. No text, captions or
watermarks anywhere.

Create ONE FINAL photorealistic live-action film still of the moment below — contemporary South Korea, 2026; all people are Korean unless stated. TIME OF DAY (lock): night.

SHOT TEXT (authoritative, Korean): 계단 아래에 서서 초점 없는 눈으로 허공을 응시하는 수리영의 정면 클로즈업

LOCATION (lock): In the villa's courtyard at night, near the foot of the exterior stairs leading to the rooftop apartment. The shot takes place here — the attached LOCATION PHOTOGRAPH shows the exact spot.

CAMERA & FRAME (follow exactly — this is the composition authority for this still; the reference images supply identity and place, never the framing):
- CAMERA: At eye level directly in front of 수리영, the camera holds an uncompromising close-up with her unfocused eyes just above center and only a narrow suggestion of the stair area behind her. Her face is frontal for this moment of psychological rupture, but her gaze passes through the lens rather than meeting it.
- FRAMING SCALE: close-up
- FRAME LAYOUT: 수리영 isolated at the bottom of the stairs in the middle-center of the frame, foreground.
- KEY BACKGROUND ELEMENTS: 계단 아래 공간 (No specific state established) — The staircase axis recedes behind her but remains indistinct in the close framing; used as Softly reduced background cue locating her at the foot of the descent.
Compose the frame exactly as specified above — camera angle, subject scale and screen placement. Keep true physical scale between people and background elements: fixtures and distant objects occupy only the small screen area their real size and distance dictate; never enlarge a background object into a foreground presence.

REALIZE FIGURATIVE LANGUAGE AS A LIVE-ACTION SHOT: the Korean shot
text may describe characters metaphorically, figuratively or with
exaggeration. Photograph what a real movie camera would actually
record on a physical set — exaggerated or figurative impressions
become realistic staging choices (lighting, distance, angle,
wardrobe), not literal fantasy imagery.
EVERY CHARACTER IS A HUMAN BEING: unless the story explicitly
features non-human or virtual beings (as in science fiction or
fantasy), every character — however indirectly, vaguely or
figuratively the text describes them — IS a real human. When the
text gives no direct visual description of a person, IMAGINE one
and still show them as a concrete, fully-formed human being:
always render the human form to the maximum extent — build,
posture, face, hands, clothing — never reduce a person to a shape,
blob, solid silhouette or abstract mass.

EXPRESSIONS ARE ACTED, NEVER ANATOMICAL: when the text describes a
person's eyes, face or presence emotionally or figuratively (vacant,
hollow, dazed, burning, lifeless gaze and the like), it describes an
ACTOR'S PERFORMANCE captured by a real camera — realize it ONLY
through gaze direction, focus, eyelids, facial muscles, stillness
and posture. Every living person's eyes remain anatomically normal
human eyes with a natural iris and pupil, natural sclera and normal
proportions; NEVER whiten, blank out, cloud over, glow, enlarge or
otherwise alter eyeballs, skin or anatomy — unless the story
explicitly declares that being non-human or supernatural in form.

PROPS FACE THE RIGHT WAY: every handheld or used object must be
oriented exactly as its real-world use requires. A person reading,
watching or operating something (a phone, a photograph, a paper,
any device) has its functional side — screen, front, page — facing
THEIR OWN eyes; the camera then sees whatever side the staging
geometry implies (often its back). Show the functional side to the
camera ONLY when the shot text itself stages it toward the viewer.
Never flip, mirror or reverse an object's front and back.

PEOPLE: the SHOT TEXT alone decides whether any person is visible in this shot. IF a person appears, they must be one of: 수리영 (한국인 20대 초반 여성, 긴 검은 생머리, 젊은 타원형 얼굴) — never anyone else, and never add a person the shot text does not show. IF only part of a person is in frame (a hand, arm, foot, back, silhouette), that body part belongs to the specific person the shot text names — its sex, age, build, skin and grooming must unmistakably match that person's profile above.

No text, captions, watermarks or annotations anywhere.

Stage the shot. The FIRST attached image (SHOT BACKGROUND) is the finished empty background of this shot — keep it EXACTLY: its camera, perspective, architecture, lighting and every fixed feature stay untouched. The SECOND attached image (LAYOUT SKETCH) tells you ONLY where the people go: each sketched person's position, screen size, pose and the gaze/motion arrows. Ignore the sketch's background lines. The CHARACTER REFERENCE photographs show the real people.
Place the real people into the background at exactly the sketched positions, sizes and poses, following the arrow directions. No sketch lines or arrows may remain.

Create ONE FINAL photorealistic live-action film still of the moment below — contemporary South Korea, 2026; all people are Korean unless stated. TIME OF DAY (lock): night.

SHOT TEXT (authoritative, Korean): 계단 아래에 서서 초점 없는 눈으로 허공을 응시하는 수리영의 정면 클로즈업

LOCATION (lock): In the villa's courtyard at night, near the foot of the exterior stairs leading to the rooftop apartment. The shot takes place here — the attached LOCATION PHOTOGRAPH shows the exact spot.

CAMERA & FRAME (follow exactly — this is the composition authority for this still; the reference images supply identity and place, never the framing):
- CAMERA: At eye level directly in front of 수리영, the camera holds an uncompromising close-up with her unfocused eyes just above center and only a narrow suggestion of the stair area behind her. Her face is frontal for this moment of psychological rupture, but her gaze passes through the lens rather than meeting it.
- FRAMING SCALE: close-up
- FRAME LAYOUT: 수리영 isolated at the bottom of the stairs in the middle-center of the frame, foreground.
- KEY BACKGROUND ELEMENTS: 계단 아래 공간 (No specific state established) — The staircase axis recedes behind her but remains indistinct in the close framing; used as Softly reduced background cue locating her at the foot of the descent.
Compose the frame exactly as specified above — camera angle, subject scale and screen placement. Keep true physical scale between people and background elements: fixtures and distant objects occupy only the small screen area their real size and distance dictate; never enlarge a background object into a foreground presence.

REALIZE FIGURATIVE LANGUAGE AS A LIVE-ACTION SHOT: the Korean shot
text may describe characters metaphorically, figuratively or with
exaggeration. Photograph what a real movie camera would actually
record on a physical set — exaggerated or figurative impressions
become realistic staging choices (lighting, distance, angle,
wardrobe), not literal fantasy imagery.
EVERY CHARACTER IS A HUMAN BEING: unless the story explicitly
features non-human or virtual beings (as in science fiction or
fantasy), every character — however indirectly, vaguely or
figuratively the text describes them — IS a real human. When the
text gives no direct visual description of a person, IMAGINE one
and still show them as a concrete, fully-formed human being:
always render the human form to the maximum extent — build,
posture, face, hands, clothing — never reduce a person to a shape,
blob, solid silhouette or abstract mass.

EXPRESSIONS ARE ACTED, NEVER ANATOMICAL: when the text describes a
person's eyes, face or presence emotionally or figuratively (vacant,
hollow, dazed, burning, lifeless gaze and the like), it describes an
ACTOR'S PERFORMANCE captured by a real camera — realize it ONLY
through gaze direction, focus, eyelids, facial muscles, stillness
and posture. Every living person's eyes remain anatomically normal
human eyes with a natural iris and pupil, natural sclera and normal
proportions; NEVER whiten, blank out, cloud over, glow, enlarge or
otherwise alter eyeballs, skin or anatomy — unless the story
explicitly declares that being non-human or supernatural in form.

PROPS FACE THE RIGHT WAY: every handheld or used object must be
oriented exactly as its real-world use requires. A person reading,
watching or operating something (a phone, a photograph, a paper,
any device) has its functional side — screen, front, page — facing
THEIR OWN eyes; the camera then sees whatever side the staging
geometry implies (often its back). Show the functional side to the
camera ONLY when the shot text itself stages it toward the viewer.
Never flip, mirror or reverse an object's front and back.

PEOPLE: the SHOT TEXT alone decides whether any person is visible in this shot. IF a person appears, they must be one of: 수리영 (한국인 20대 초반 여성, 긴 검은 생머리, 젊은 타원형 얼굴) — never anyone else, and never add a person the shot text does not show. IF only part of a person is in frame (a hand, arm, foot, back, silhouette), that body part belongs to the specific person the shot text names — its sex, age, build, skin and grooming must unmistakably match that person's profile above.

No text, captions, watermarks or annotations anywhere.

Create ONE FINAL photorealistic live-action film still of the moment below — contemporary South Korea, 2026; all people are Korean unless stated. TIME OF DAY (lock): night.

SHOT TEXT (authoritative, Korean): 계단 아래에 서서 초점 없는 눈으로 허공을 응시하는 수리영의 정면 클로즈업

LOCATION (lock): In the villa's courtyard at night, near the foot of the exterior stairs leading to the rooftop apartment. The shot takes place here — the attached LOCATION PHOTOGRAPH shows the exact spot.

CAMERA & FRAME (follow exactly — this is the composition authority for this still; the reference images supply identity and place, never the framing):
- CAMERA: At eye level directly in front of 수리영, the camera holds an uncompromising close-up with her unfocused eyes just above center and only a narrow suggestion of the stair area behind her. Her face is frontal for this moment of psychological rupture, but her gaze passes through the lens rather than meeting it.
- FRAMING SCALE: close-up
- FRAME LAYOUT: 수리영 isolated at the bottom of the stairs in the middle-center of the frame, foreground.
- KEY BACKGROUND ELEMENTS: 계단 아래 공간 (No specific state established) — The staircase axis recedes behind her but remains indistinct in the close framing; used as Softly reduced background cue locating her at the foot of the descent.
Compose the frame exactly as specified above — camera angle, subject scale and screen placement. Keep true physical scale between people and background elements: fixtures and distant objects occupy only the small screen area their real size and distance dictate; never enlarge a background object into a foreground presence.

REALIZE FIGURATIVE LANGUAGE AS A LIVE-ACTION SHOT: the Korean shot
text may describe characters metaphorically, figuratively or with
exaggeration. Photograph what a real movie camera would actually
record on a physical set — exaggerated or figurative impressions
become realistic staging choices (lighting, distance, angle,
wardrobe), not literal fantasy imagery.
EVERY CHARACTER IS A HUMAN BEING: unless the story explicitly
features non-human or virtual beings (as in science fiction or
fantasy), every character — however indirectly, vaguely or
figuratively the text describes them — IS a real human. When the
text gives no direct visual description of a person, IMAGINE one
and still show them as a concrete, fully-formed human being:
always render the human form to the maximum extent — build,
posture, face, hands, clothing — never reduce a person to a shape,
blob, solid silhouette or abstract mass.

EXPRESSIONS ARE ACTED, NEVER ANATOMICAL: when the text describes a
person's eyes, face or presence emotionally or figuratively (vacant,
hollow, dazed, burning, lifeless gaze and the like), it describes an
ACTOR'S PERFORMANCE captured by a real camera — realize it ONLY
through gaze direction, focus, eyelids, facial muscles, stillness
and posture. Every living person's eyes remain anatomically normal
human eyes with a natural iris and pupil, natural sclera and normal
proportions; NEVER whiten, blank out, cloud over, glow, enlarge or
otherwise alter eyeballs, skin or anatomy — unless the story
explicitly declares that being non-human or supernatural in form.

PROPS FACE THE RIGHT WAY: every handheld or used object must be
oriented exactly as its real-world use requires. A person reading,
watching or operating something (a phone, a photograph, a paper,
any device) has its functional side — screen, front, page — facing
THEIR OWN eyes; the camera then sees whatever side the staging
geometry implies (often its back). Show the functional side to the
camera ONLY when the shot text itself stages it toward the viewer.
Never flip, mirror or reverse an object's front and back.

PEOPLE: the SHOT TEXT alone decides whether any person is visible in this shot. IF a person appears, they must be one of: 수리영 (한국인 20대 초반 여성, 긴 검은 생머리, 젊은 타원형 얼굴) — never anyone else, and never add a person the shot text does not show. IF only part of a person is in frame (a hand, arm, foot, back, silhouette), that body part belongs to the specific person the shot text names — its sex, age, build, skin and grooming must unmistakably match that person's profile above.

No text, captions, watermarks or annotations anywhere.

Edit this photograph. Apply ONLY the corrections listed below,
each in place, changing nothing else:

CORRECTIONS:
- Change the shot scale to a tight close-up on her face, and heavily blur the background to leave only a narrow, indistinct suggestion of the stairs.
- Add the olive-green baseball cap from the reference image onto her head.

PRESERVE EVERYTHING ELSE EXACTLY as in the original photograph —
every person, pose, face, garment, object, set feature, lighting,
camera and framing stay identical. No text, captions or
watermarks anywhere.

Create ONE FINAL photorealistic live-action film still of the moment below — contemporary South Korea, 2026; all people are Korean unless stated. TIME OF DAY (lock): night.

SHOT TEXT (authoritative, Korean): 경찰서 열린 문간에 서서 귀에 휴대폰을 댄 채 굳은 표정으로 멈춰 있는 혜수의 상체

LOCATION (lock): At the open doorway leading into the police office at night, with the duty room beyond lit for work. The shot takes place here — the attached LOCATION PHOTOGRAPH shows the exact spot.

CAMERA & FRAME (follow exactly — this is the composition authority for this still; the reference images supply identity and place, never the framing):
- CAMERA: From just inside the police office at chest height, frame the open doorway in a medium shot as 혜수 pauses on its threshold, the phone pressed to her ear and her upper body held in three-quarter view. Leave open office space behind the camera-facing side of the threshold so her eventual movement toward 김형사 has a clear path for the backward track.
- FRAMING SCALE: medium shot
- KEY BACKGROUND ELEMENTS: office doorway (Open) — The inner sides of the doorway face the camera from within the office; used as Frames 혜수 at the boundary between entry and office space; police office interior (Occupied only by 김형사 before 혜수 enters); used as Provides clear depth behind the threshold and the route for 혜수's entrance.
Compose the frame exactly as specified above — camera angle, subject scale and screen placement. Keep true physical scale between people and background elements: fixtures and distant objects occupy only the small screen area their real size and distance dictate; never enlarge a background object into a foreground presence.

REALIZE FIGURATIVE LANGUAGE AS A LIVE-ACTION SHOT: the Korean shot
text may describe characters metaphorically, figuratively or with
exaggeration. Photograph what a real movie camera would actually
record on a physical set — exaggerated or figurative impressions
become realistic staging choices (lighting, distance, angle,
wardrobe), not literal fantasy imagery.
EVERY CHARACTER IS A HUMAN BEING: unless the story explicitly
features non-human or virtual beings (as in science fiction or
fantasy), every character — however indirectly, vaguely or
figuratively the text describes them — IS a real human. When the
text gives no direct visual description of a person, IMAGINE one
and still show them as a concrete, fully-formed human being:
always render the human form to the maximum extent — build,
posture, face, hands, clothing — never reduce a person to a shape,
blob, solid silhouette or abstract mass.

EXPRESSIONS ARE ACTED, NEVER ANATOMICAL: when the text describes a
person's eyes, face or presence emotionally or figuratively (vacant,
hollow, dazed, burning, lifeless gaze and the like), it describes an
ACTOR'S PERFORMANCE captured by a real camera — realize it ONLY
through gaze direction, focus, eyelids, facial muscles, stillness
and posture. Every living person's eyes remain anatomically normal
human eyes with a natural iris and pupil, natural sclera and normal
proportions; NEVER whiten, blank out, cloud over, glow, enlarge or
otherwise alter eyeballs, skin or anatomy — unless the story
explicitly declares that being non-human or supernatural in form.

PROPS FACE THE RIGHT WAY: every handheld or used object must be
oriented exactly as its real-world use requires. A person reading,
watching or operating something (a phone, a photograph, a paper,
any device) has its functional side — screen, front, page — facing
THEIR OWN eyes; the camera then sees whatever side the staging
geometry implies (often its back). Show the functional side to the
camera ONLY when the shot text itself stages it toward the viewer.
Never flip, mirror or reverse an object's front and back.

CARRIED STATE (persist exactly — must match the neighbouring shots of this scene): Hye-su has her phone to her ear, and Min-suk’s number continues to give no response.

PEOPLE: the SHOT TEXT alone decides whether any person is visible in this shot. IF a person appears, they must be one of: 혜수 (한국인 40대 초반 여성, 단정한 검은 머리, 성숙한 타원형 얼굴) — never anyone else, and never add a person the shot text does not show. IF only part of a person is in frame (a hand, arm, foot, back, silhouette), that body part belongs to the specific person the shot text names — its sex, age, build, skin and grooming must unmistakably match that person's profile above.

No text, captions, watermarks or annotations anywhere.

Stage the shot. The FIRST attached image (SHOT BACKGROUND) is the finished empty background of this shot — keep it EXACTLY: its camera, perspective, architecture, lighting and every fixed feature stay untouched. The SECOND attached image (LAYOUT SKETCH) tells you ONLY where the people go: each sketched person's position, screen size, pose and the gaze/motion arrows. Ignore the sketch's background lines. The CHARACTER REFERENCE photographs show the real people.
Place the real people into the background at exactly the sketched positions, sizes and poses, following the arrow directions. No sketch lines or arrows may remain.

Create ONE FINAL photorealistic live-action film still of the moment below — contemporary South Korea, 2026; all people are Korean unless stated. TIME OF DAY (lock): night.

SHOT TEXT (authoritative, Korean): 경찰서 열린 문간에 서서 귀에 휴대폰을 댄 채 굳은 표정으로 멈춰 있는 혜수의 상체

LOCATION (lock): At the open doorway leading into the police office at night, with the duty room beyond lit for work. The shot takes place here — the attached LOCATION PHOTOGRAPH shows the exact spot.

CAMERA & FRAME (follow exactly — this is the composition authority for this still; the reference images supply identity and place, never the framing):
- CAMERA: From just inside the police office at chest height, frame the open doorway in a medium shot as 혜수 pauses on its threshold, the phone pressed to her ear and her upper body held in three-quarter view. Leave open office space behind the camera-facing side of the threshold so her eventual movement toward 김형사 has a clear path for the backward track.
- FRAMING SCALE: medium shot
- KEY BACKGROUND ELEMENTS: office doorway (Open) — The inner sides of the doorway face the camera from within the office; used as Frames 혜수 at the boundary between entry and office space; police office interior (Occupied only by 김형사 before 혜수 enters); used as Provides clear depth behind the threshold and the route for 혜수's entrance.
Compose the frame exactly as specified above — camera angle, subject scale and screen placement. Keep true physical scale between people and background elements: fixtures and distant objects occupy only the small screen area their real size and distance dictate; never enlarge a background object into a foreground presence.

REALIZE FIGURATIVE LANGUAGE AS A LIVE-ACTION SHOT: the Korean shot
text may describe characters metaphorically, figuratively or with
exaggeration. Photograph what a real movie camera would actually
record on a physical set — exaggerated or figurative impressions
become realistic staging choices (lighting, distance, angle,
wardrobe), not literal fantasy imagery.
EVERY CHARACTER IS A HUMAN BEING: unless the story explicitly
features non-human or virtual beings (as in science fiction or
fantasy), every character — however indirectly, vaguely or
figuratively the text describes them — IS a real human. When the
text gives no direct visual description of a person, IMAGINE one
and still show them as a concrete, fully-formed human being:
always render the human form to the maximum extent — build,
posture, face, hands, clothing — never reduce a person to a shape,
blob, solid silhouette or abstract mass.

EXPRESSIONS ARE ACTED, NEVER ANATOMICAL: when the text describes a
person's eyes, face or presence emotionally or figuratively (vacant,
hollow, dazed, burning, lifeless gaze and the like), it describes an
ACTOR'S PERFORMANCE captured by a real camera — realize it ONLY
through gaze direction, focus, eyelids, facial muscles, stillness
and posture. Every living person's eyes remain anatomically normal
human eyes with a natural iris and pupil, natural sclera and normal
proportions; NEVER whiten, blank out, cloud over, glow, enlarge or
otherwise alter eyeballs, skin or anatomy — unless the story
explicitly declares that being non-human or supernatural in form.

PROPS FACE THE RIGHT WAY: every handheld or used object must be
oriented exactly as its real-world use requires. A person reading,
watching or operating something (a phone, a photograph, a paper,
any device) has its functional side — screen, front, page — facing
THEIR OWN eyes; the camera then sees whatever side the staging
geometry implies (often its back). Show the functional side to the
camera ONLY when the shot text itself stages it toward the viewer.
Never flip, mirror or reverse an object's front and back.

CARRIED STATE (persist exactly — must match the neighbouring shots of this scene): Hye-su has her phone to her ear, and Min-suk’s number continues to give no response.

PEOPLE: the SHOT TEXT alone decides whether any person is visible in this shot. IF a person appears, they must be one of: 혜수 (한국인 40대 초반 여성, 단정한 검은 머리, 성숙한 타원형 얼굴) — never anyone else, and never add a person the shot text does not show. IF only part of a person is in frame (a hand, arm, foot, back, silhouette), that body part belongs to the specific person the shot text names — its sex, age, build, skin and grooming must unmistakably match that person's profile above.

No text, captions, watermarks or annotations anywhere.

Create ONE FINAL photorealistic live-action film still of the moment below — contemporary South Korea, 2026; all people are Korean unless stated. TIME OF DAY (lock): night.

SHOT TEXT (authoritative, Korean): 경찰서 열린 문간에 서서 귀에 휴대폰을 댄 채 굳은 표정으로 멈춰 있는 혜수의 상체

LOCATION (lock): At the open doorway leading into the police office at night, with the duty room beyond lit for work. The shot takes place here — the attached LOCATION PHOTOGRAPH shows the exact spot.

CAMERA & FRAME (follow exactly — this is the composition authority for this still; the reference images supply identity and place, never the framing):
- CAMERA: From just inside the police office at chest height, frame the open doorway in a medium shot as 혜수 pauses on its threshold, the phone pressed to her ear and her upper body held in three-quarter view. Leave open office space behind the camera-facing side of the threshold so her eventual movement toward 김형사 has a clear path for the backward track.
- FRAMING SCALE: medium shot
- KEY BACKGROUND ELEMENTS: office doorway (Open) — The inner sides of the doorway face the camera from within the office; used as Frames 혜수 at the boundary between entry and office space; police office interior (Occupied only by 김형사 before 혜수 enters); used as Provides clear depth behind the threshold and the route for 혜수's entrance.
Compose the frame exactly as specified above — camera angle, subject scale and screen placement. Keep true physical scale between people and background elements: fixtures and distant objects occupy only the small screen area their real size and distance dictate; never enlarge a background object into a foreground presence.

REALIZE FIGURATIVE LANGUAGE AS A LIVE-ACTION SHOT: the Korean shot
text may describe characters metaphorically, figuratively or with
exaggeration. Photograph what a real movie camera would actually
record on a physical set — exaggerated or figurative impressions
become realistic staging choices (lighting, distance, angle,
wardrobe), not literal fantasy imagery.
EVERY CHARACTER IS A HUMAN BEING: unless the story explicitly
features non-human or virtual beings (as in science fiction or
fantasy), every character — however indirectly, vaguely or
figuratively the text describes them — IS a real human. When the
text gives no direct visual description of a person, IMAGINE one
and still show them as a concrete, fully-formed human being:
always render the human form to the maximum extent — build,
posture, face, hands, clothing — never reduce a person to a shape,
blob, solid silhouette or abstract mass.

EXPRESSIONS ARE ACTED, NEVER ANATOMICAL: when the text describes a
person's eyes, face or presence emotionally or figuratively (vacant,
hollow, dazed, burning, lifeless gaze and the like), it describes an
ACTOR'S PERFORMANCE captured by a real camera — realize it ONLY
through gaze direction, focus, eyelids, facial muscles, stillness
and posture. Every living person's eyes remain anatomically normal
human eyes with a natural iris and pupil, natural sclera and normal
proportions; NEVER whiten, blank out, cloud over, glow, enlarge or
otherwise alter eyeballs, skin or anatomy — unless the story
explicitly declares that being non-human or supernatural in form.

PROPS FACE THE RIGHT WAY: every handheld or used object must be
oriented exactly as its real-world use requires. A person reading,
watching or operating something (a phone, a photograph, a paper,
any device) has its functional side — screen, front, page — facing
THEIR OWN eyes; the camera then sees whatever side the staging
geometry implies (often its back). Show the functional side to the
camera ONLY when the shot text itself stages it toward the viewer.
Never flip, mirror or reverse an object's front and back.

CARRIED STATE (persist exactly — must match the neighbouring shots of this scene): Hye-su has her phone to her ear, and Min-suk’s number continues to give no response.

PEOPLE: the SHOT TEXT alone decides whether any person is visible in this shot. IF a person appears, they must be one of: 혜수 (한국인 40대 초반 여성, 단정한 검은 머리, 성숙한 타원형 얼굴) — never anyone else, and never add a person the shot text does not show. IF only part of a person is in frame (a hand, arm, foot, back, silhouette), that body part belongs to the specific person the shot text names — its sex, age, build, skin and grooming must unmistakably match that person's profile above.

No text, captions, watermarks or annotations anywhere.

Edit this photograph. Apply ONLY the corrections listed below,
each in place, changing nothing else:

CORRECTIONS:
- Redraw the right hand so it has distinct, anatomically correct fingers naturally gripping the phone.

PRESERVE EVERYTHING ELSE EXACTLY as in the original photograph —
every person, pose, face, garment, object, set feature, lighting,
camera and framing stay identical. No text, captions or
watermarks anywhere.

Create ONE FINAL photorealistic live-action film still of the moment below — contemporary South Korea, 2026; all people are Korean unless stated. TIME OF DAY (lock): late morning.

SHOT TEXT (authoritative, Korean): 수리영이 상체를 굽히고 자전거 페달을 강하게 밟으며 앞으로 질주하는 순간의 전신, 머리카락과 옷자락이 뒤로 날리는 상태

LOCATION (lock): On the street immediately outside the police station in daylight, accelerating away on a bicycle. The shot takes place here — the attached LOCATION PHOTOGRAPH shows the exact spot.

CAMERA & FRAME (follow exactly — this is the composition authority for this still; the reference images supply identity and place, never the framing):
- CAMERA: Track parallel at waist height from Suriyoung's lateral side, holding her full body and bicycle in a wide profile as she leans forward and drives the pedals. Keep her near the middle of frame while the travel path streams toward screen right, preserving the stage's stable side axis and emphasizing acceleration rather than changing camera angle.
- FRAMING SCALE: wide shot
- FRAME LAYOUT: 수리영 in the middle-center of the frame, midground, moves toward screen-right travel path; screen-right travel path in the middle-right of the frame, background.
- KEY BACKGROUND ELEMENTS: 자전거 (Moving and accelerating) — Its lateral side faces the camera, keeping both the frame and pedaling motion readable; used as Carries Suriyoung through the frame and supplies the visible pedaling action; 경찰서 외부 (Visible behind the moving bicycle) — The outward-facing exterior recedes behind her direction of travel; used as Background origin for her departure.
Compose the frame exactly as specified above — camera angle, subject scale and screen placement. Keep true physical scale between people and background elements: fixtures and distant objects occupy only the small screen area their real size and distance dictate; never enlarge a background object into a foreground presence.

REALIZE FIGURATIVE LANGUAGE AS A LIVE-ACTION SHOT: the Korean shot
text may describe characters metaphorically, figuratively or with
exaggeration. Photograph what a real movie camera would actually
record on a physical set — exaggerated or figurative impressions
become realistic staging choices (lighting, distance, angle,
wardrobe), not literal fantasy imagery.
EVERY CHARACTER IS A HUMAN BEING: unless the story explicitly
features non-human or virtual beings (as in science fiction or
fantasy), every character — however indirectly, vaguely or
figuratively the text describes them — IS a real human. When the
text gives no direct visual description of a person, IMAGINE one
and still show them as a concrete, fully-formed human being:
always render the human form to the maximum extent — build,
posture, face, hands, clothing — never reduce a person to a shape,
blob, solid silhouette or abstract mass.

EXPRESSIONS ARE ACTED, NEVER ANATOMICAL: when the text describes a
person's eyes, face or presence emotionally or figuratively (vacant,
hollow, dazed, burning, lifeless gaze and the like), it describes an
ACTOR'S PERFORMANCE captured by a real camera — realize it ONLY
through gaze direction, focus, eyelids, facial muscles, stillness
and posture. Every living person's eyes remain anatomically normal
human eyes with a natural iris and pupil, natural sclera and normal
proportions; NEVER whiten, blank out, cloud over, glow, enlarge or
otherwise alter eyeballs, skin or anatomy — unless the story
explicitly declares that being non-human or supernatural in form.

PROPS FACE THE RIGHT WAY: every handheld or used object must be
oriented exactly as its real-world use requires. A person reading,
watching or operating something (a phone, a photograph, a paper,
any device) has its functional side — screen, front, page — facing
THEIR OWN eyes; the camera then sees whatever side the staging
geometry implies (often its back). Show the functional side to the
camera ONLY when the shot text itself stages it toward the viewer.
Never flip, mirror or reverse an object's front and back.

CARRIED STATE (persist exactly — must match the neighbouring shots of this scene): Su-ri-young rides away on her bicycle with the folded coordinate note still in her pocket and her doll-adorned backpack and phone still with her.

PEOPLE: the SHOT TEXT alone decides whether any person is visible in this shot. IF a person appears, they must be one of: 수리영 (한국인 20대 초반 여성, 긴 검은 생머리, 젊은 타원형 얼굴) — never anyone else, and never add a person the shot text does not show. IF only part of a person is in frame (a hand, arm, foot, back, silhouette), that body part belongs to the specific person the shot text names — its sex, age, build, skin and grooming must unmistakably match that person's profile above.

No text, captions, watermarks or annotations anywhere.

Stage the shot. The FIRST attached image (SHOT BACKGROUND) is the finished empty background of this shot — keep it EXACTLY: its camera, perspective, architecture, lighting and every fixed feature stay untouched. The SECOND attached image (LAYOUT SKETCH) tells you ONLY where the people go: each sketched person's position, screen size, pose and the gaze/motion arrows. Ignore the sketch's background lines. The CHARACTER REFERENCE photographs show the real people.
Place the real people into the background at exactly the sketched positions, sizes and poses, following the arrow directions. No sketch lines or arrows may remain.

Create ONE FINAL photorealistic live-action film still of the moment below — contemporary South Korea, 2026; all people are Korean unless stated. TIME OF DAY (lock): late morning.

SHOT TEXT (authoritative, Korean): 수리영이 상체를 굽히고 자전거 페달을 강하게 밟으며 앞으로 질주하는 순간의 전신, 머리카락과 옷자락이 뒤로 날리는 상태

LOCATION (lock): On the street immediately outside the police station in daylight, accelerating away on a bicycle. The shot takes place here — the attached LOCATION PHOTOGRAPH shows the exact spot.

CAMERA & FRAME (follow exactly — this is the composition authority for this still; the reference images supply identity and place, never the framing):
- CAMERA: Track parallel at waist height from Suriyoung's lateral side, holding her full body and bicycle in a wide profile as she leans forward and drives the pedals. Keep her near the middle of frame while the travel path streams toward screen right, preserving the stage's stable side axis and emphasizing acceleration rather than changing camera angle.
- FRAMING SCALE: wide shot
- FRAME LAYOUT: 수리영 in the middle-center of the frame, midground, moves toward screen-right travel path; screen-right travel path in the middle-right of the frame, background.
- KEY BACKGROUND ELEMENTS: 자전거 (Moving and accelerating) — Its lateral side faces the camera, keeping both the frame and pedaling motion readable; used as Carries Suriyoung through the frame and supplies the visible pedaling action; 경찰서 외부 (Visible behind the moving bicycle) — The outward-facing exterior recedes behind her direction of travel; used as Background origin for her departure.
Compose the frame exactly as specified above — camera angle, subject scale and screen placement. Keep true physical scale between people and background elements: fixtures and distant objects occupy only the small screen area their real size and distance dictate; never enlarge a background object into a foreground presence.

REALIZE FIGURATIVE LANGUAGE AS A LIVE-ACTION SHOT: the Korean shot
text may describe characters metaphorically, figuratively or with
exaggeration. Photograph what a real movie camera would actually
record on a physical set — exaggerated or figurative impressions
become realistic staging choices (lighting, distance, angle,
wardrobe), not literal fantasy imagery.
EVERY CHARACTER IS A HUMAN BEING: unless the story explicitly
features non-human or virtual beings (as in science fiction or
fantasy), every character — however indirectly, vaguely or
figuratively the text describes them — IS a real human. When the
text gives no direct visual description of a person, IMAGINE one
and still show them as a concrete, fully-formed human being:
always render the human form to the maximum extent — build,
posture, face, hands, clothing — never reduce a person to a shape,
blob, solid silhouette or abstract mass.

EXPRESSIONS ARE ACTED, NEVER ANATOMICAL: when the text describes a
person's eyes, face or presence emotionally or figuratively (vacant,
hollow, dazed, burning, lifeless gaze and the like), it describes an
ACTOR'S PERFORMANCE captured by a real camera — realize it ONLY
through gaze direction, focus, eyelids, facial muscles, stillness
and posture. Every living person's eyes remain anatomically normal
human eyes with a natural iris and pupil, natural sclera and normal
proportions; NEVER whiten, blank out, cloud over, glow, enlarge or
otherwise alter eyeballs, skin or anatomy — unless the story
explicitly declares that being non-human or supernatural in form.

PROPS FACE THE RIGHT WAY: every handheld or used object must be
oriented exactly as its real-world use requires. A person reading,
watching or operating something (a phone, a photograph, a paper,
any device) has its functional side — screen, front, page — facing
THEIR OWN eyes; the camera then sees whatever side the staging
geometry implies (often its back). Show the functional side to the
camera ONLY when the shot text itself stages it toward the viewer.
Never flip, mirror or reverse an object's front and back.

CARRIED STATE (persist exactly — must match the neighbouring shots of this scene): Su-ri-young rides away on her bicycle with the folded coordinate note still in her pocket and her doll-adorned backpack and phone still with her.

PEOPLE: the SHOT TEXT alone decides whether any person is visible in this shot. IF a person appears, they must be one of: 수리영 (한국인 20대 초반 여성, 긴 검은 생머리, 젊은 타원형 얼굴) — never anyone else, and never add a person the shot text does not show. IF only part of a person is in frame (a hand, arm, foot, back, silhouette), that body part belongs to the specific person the shot text names — its sex, age, build, skin and grooming must unmistakably match that person's profile above.

No text, captions, watermarks or annotations anywhere.

Create ONE FINAL photorealistic live-action film still of the moment below — contemporary South Korea, 2026; all people are Korean unless stated. TIME OF DAY (lock): late morning.

SHOT TEXT (authoritative, Korean): 수리영이 상체를 굽히고 자전거 페달을 강하게 밟으며 앞으로 질주하는 순간의 전신, 머리카락과 옷자락이 뒤로 날리는 상태

LOCATION (lock): On the street immediately outside the police station in daylight, accelerating away on a bicycle. The shot takes place here — the attached LOCATION PHOTOGRAPH shows the exact spot.

CAMERA & FRAME (follow exactly — this is the composition authority for this still; the reference images supply identity and place, never the framing):
- CAMERA: Track parallel at waist height from Suriyoung's lateral side, holding her full body and bicycle in a wide profile as she leans forward and drives the pedals. Keep her near the middle of frame while the travel path streams toward screen right, preserving the stage's stable side axis and emphasizing acceleration rather than changing camera angle.
- FRAMING SCALE: wide shot
- FRAME LAYOUT: 수리영 in the middle-center of the frame, midground, moves toward screen-right travel path; screen-right travel path in the middle-right of the frame, background.
- KEY BACKGROUND ELEMENTS: 자전거 (Moving and accelerating) — Its lateral side faces the camera, keeping both the frame and pedaling motion readable; used as Carries Suriyoung through the frame and supplies the visible pedaling action; 경찰서 외부 (Visible behind the moving bicycle) — The outward-facing exterior recedes behind her direction of travel; used as Background origin for her departure.
Compose the frame exactly as specified above — camera angle, subject scale and screen placement. Keep true physical scale between people and background elements: fixtures and distant objects occupy only the small screen area their real size and distance dictate; never enlarge a background object into a foreground presence.

REALIZE FIGURATIVE LANGUAGE AS A LIVE-ACTION SHOT: the Korean shot
text may describe characters metaphorically, figuratively or with
exaggeration. Photograph what a real movie camera would actually
record on a physical set — exaggerated or figurative impressions
become realistic staging choices (lighting, distance, angle,
wardrobe), not literal fantasy imagery.
EVERY CHARACTER IS A HUMAN BEING: unless the story explicitly
features non-human or virtual beings (as in science fiction or
fantasy), every character — however indirectly, vaguely or
figuratively the text describes them — IS a real human. When the
text gives no direct visual description of a person, IMAGINE one
and still show them as a concrete, fully-formed human being:
always render the human form to the maximum extent — build,
posture, face, hands, clothing — never reduce a person to a shape,
blob, solid silhouette or abstract mass.

EXPRESSIONS ARE ACTED, NEVER ANATOMICAL: when the text describes a
person's eyes, face or presence emotionally or figuratively (vacant,
hollow, dazed, burning, lifeless gaze and the like), it describes an
ACTOR'S PERFORMANCE captured by a real camera — realize it ONLY
through gaze direction, focus, eyelids, facial muscles, stillness
and posture. Every living person's eyes remain anatomically normal
human eyes with a natural iris and pupil, natural sclera and normal
proportions; NEVER whiten, blank out, cloud over, glow, enlarge or
otherwise alter eyeballs, skin or anatomy — unless the story
explicitly declares that being non-human or supernatural in form.

PROPS FACE THE RIGHT WAY: every handheld or used object must be
oriented exactly as its real-world use requires. A person reading,
watching or operating something (a phone, a photograph, a paper,
any device) has its functional side — screen, front, page — facing
THEIR OWN eyes; the camera then sees whatever side the staging
geometry implies (often its back). Show the functional side to the
camera ONLY when the shot text itself stages it toward the viewer.
Never flip, mirror or reverse an object's front and back.

CARRIED STATE (persist exactly — must match the neighbouring shots of this scene): Su-ri-young rides away on her bicycle with the folded coordinate note still in her pocket and her doll-adorned backpack and phone still with her.

PEOPLE: the SHOT TEXT alone decides whether any person is visible in this shot. IF a person appears, they must be one of: 수리영 (한국인 20대 초반 여성, 긴 검은 생머리, 젊은 타원형 얼굴) — never anyone else, and never add a person the shot text does not show. IF only part of a person is in frame (a hand, arm, foot, back, silhouette), that body part belongs to the specific person the shot text names — its sex, age, build, skin and grooming must unmistakably match that person's profile above.

No text, captions, watermarks or annotations anywhere.

Edit this photograph. Apply ONLY the corrections listed below,
each in place, changing nothing else:

CORRECTIONS:
- Add the character's left leg extending downwards to rest on the lower pedal behind the bicycle frame.

PRESERVE EVERYTHING ELSE EXACTLY as in the original photograph —
every person, pose, face, garment, object, set feature, lighting,
camera and framing stay identical. No text, captions or
watermarks anywhere.

Create ONE FINAL photorealistic live-action film still of the moment below — contemporary South Korea, 2026; all people are Korean unless stated. TIME OF DAY (lock): dusk.

SHOT TEXT (authoritative, Korean): 인우의 두 손이 조타실의 나무 타륜을 꽉 쥐고 팔뚝에 힘줄이 불거진 손 클로즈업

LOCATION (lock): Inside the small fishing boat's wheelhouse at sunset, directly at the wooden helm. The shot takes place here — the attached LOCATION PHOTOGRAPH shows the exact spot.

CAMERA & FRAME (follow exactly — this is the composition authority for this still; the reference images supply identity and place, never the framing):
- CAMERA: At wheel height inside the wheelhouse, the camera completes its tilt across 인우's tensed forearms and settles on both hands clamped around the wooden wheel. The hands and forearms occupy the central two-thirds, with the wheel rim and spokes held below the nonhuman-object occupancy limit and the surrounding wheelhouse reduced to narrow edge context.
- FRAMING SCALE: close-up
- KEY BACKGROUND ELEMENTS: wooden ship's wheel (Gripped tightly by both hands) — The rim and spokes are seen obliquely from wheel height, crossing behind both forearms; used as Primary tactile anchor beneath 인우's hands, retaining enough visible rim and spokes to identify the steering action; wheelhouse interior (Visible only at the frame edges); used as Narrow peripheral context that establishes the shot inside the wheelhouse.
Compose the frame exactly as specified above — camera angle, subject scale and screen placement. Keep true physical scale between people and background elements: fixtures and distant objects occupy only the small screen area their real size and distance dictate; never enlarge a background object into a foreground presence.

REALIZE FIGURATIVE LANGUAGE AS A LIVE-ACTION SHOT: the Korean shot
text may describe characters metaphorically, figuratively or with
exaggeration. Photograph what a real movie camera would actually
record on a physical set — exaggerated or figurative impressions
become realistic staging choices (lighting, distance, angle,
wardrobe), not literal fantasy imagery.
EVERY CHARACTER IS A HUMAN BEING: unless the story explicitly
features non-human or virtual beings (as in science fiction or
fantasy), every character — however indirectly, vaguely or
figuratively the text describes them — IS a real human. When the
text gives no direct visual description of a person, IMAGINE one
and still show them as a concrete, fully-formed human being:
always render the human form to the maximum extent — build,
posture, face, hands, clothing — never reduce a person to a shape,
blob, solid silhouette or abstract mass.

EXPRESSIONS ARE ACTED, NEVER ANATOMICAL: when the text describes a
person's eyes, face or presence emotionally or figuratively (vacant,
hollow, dazed, burning, lifeless gaze and the like), it describes an
ACTOR'S PERFORMANCE captured by a real camera — realize it ONLY
through gaze direction, focus, eyelids, facial muscles, stillness
and posture. Every living person's eyes remain anatomically normal
human eyes with a natural iris and pupil, natural sclera and normal
proportions; NEVER whiten, blank out, cloud over, glow, enlarge or
otherwise alter eyeballs, skin or anatomy — unless the story
explicitly declares that being non-human or supernatural in form.

PROPS FACE THE RIGHT WAY: every handheld or used object must be
oriented exactly as its real-world use requires. A person reading,
watching or operating something (a phone, a photograph, a paper,
any device) has its functional side — screen, front, page — facing
THEIR OWN eyes; the camera then sees whatever side the staging
geometry implies (often its back). Show the functional side to the
camera ONLY when the shot text itself stages it toward the viewer.
Never flip, mirror or reverse an object's front and back.

CARRIED STATE (persist exactly — must match the neighbouring shots of this scene): In-woo is trying to turn the same boat back from inside the wheelhouse. Su-ri-young follows with the old photograph, while her backpack, maps, coordinate note, and phone remain aboard.

PEOPLE: the SHOT TEXT alone decides whether any person is visible in this shot. IF a person appears, they must be one of: 인우 (한국인 20대 중반 남성, 짧은 검은 머리, 반듯한 얼굴형, 젊은 인상) — never anyone else, and never add a person the shot text does not show. IF only part of a person is in frame (a hand, arm, foot, back, silhouette), that body part belongs to the specific person the shot text names — its sex, age, build, skin and grooming must unmistakably match that person's profile above.

No text, captions, watermarks or annotations anywhere.

Stage the shot. The FIRST attached image (SHOT BACKGROUND) is the finished empty background of this shot — keep it EXACTLY: its camera, perspective, architecture, lighting and every fixed feature stay untouched. The SECOND attached image (LAYOUT SKETCH) tells you ONLY where the people go: each sketched person's position, screen size, pose and the gaze/motion arrows. Ignore the sketch's background lines. The CHARACTER REFERENCE photographs show the real people.
Place the real people into the background at exactly the sketched positions, sizes and poses, following the arrow directions. No sketch lines or arrows may remain.

Create ONE FINAL photorealistic live-action film still of the moment below — contemporary South Korea, 2026; all people are Korean unless stated. TIME OF DAY (lock): dusk.

SHOT TEXT (authoritative, Korean): 인우의 두 손이 조타실의 나무 타륜을 꽉 쥐고 팔뚝에 힘줄이 불거진 손 클로즈업

LOCATION (lock): Inside the small fishing boat's wheelhouse at sunset, directly at the wooden helm. The shot takes place here — the attached LOCATION PHOTOGRAPH shows the exact spot.

CAMERA & FRAME (follow exactly — this is the composition authority for this still; the reference images supply identity and place, never the framing):
- CAMERA: At wheel height inside the wheelhouse, the camera completes its tilt across 인우's tensed forearms and settles on both hands clamped around the wooden wheel. The hands and forearms occupy the central two-thirds, with the wheel rim and spokes held below the nonhuman-object occupancy limit and the surrounding wheelhouse reduced to narrow edge context.
- FRAMING SCALE: close-up
- KEY BACKGROUND ELEMENTS: wooden ship's wheel (Gripped tightly by both hands) — The rim and spokes are seen obliquely from wheel height, crossing behind both forearms; used as Primary tactile anchor beneath 인우's hands, retaining enough visible rim and spokes to identify the steering action; wheelhouse interior (Visible only at the frame edges); used as Narrow peripheral context that establishes the shot inside the wheelhouse.
Compose the frame exactly as specified above — camera angle, subject scale and screen placement. Keep true physical scale between people and background elements: fixtures and distant objects occupy only the small screen area their real size and distance dictate; never enlarge a background object into a foreground presence.

REALIZE FIGURATIVE LANGUAGE AS A LIVE-ACTION SHOT: the Korean shot
text may describe characters metaphorically, figuratively or with
exaggeration. Photograph what a real movie camera would actually
record on a physical set — exaggerated or figurative impressions
become realistic staging choices (lighting, distance, angle,
wardrobe), not literal fantasy imagery.
EVERY CHARACTER IS A HUMAN BEING: unless the story explicitly
features non-human or virtual beings (as in science fiction or
fantasy), every character — however indirectly, vaguely or
figuratively the text describes them — IS a real human. When the
text gives no direct visual description of a person, IMAGINE one
and still show them as a concrete, fully-formed human being:
always render the human form to the maximum extent — build,
posture, face, hands, clothing — never reduce a person to a shape,
blob, solid silhouette or abstract mass.

EXPRESSIONS ARE ACTED, NEVER ANATOMICAL: when the text describes a
person's eyes, face or presence emotionally or figuratively (vacant,
hollow, dazed, burning, lifeless gaze and the like), it describes an
ACTOR'S PERFORMANCE captured by a real camera — realize it ONLY
through gaze direction, focus, eyelids, facial muscles, stillness
and posture. Every living person's eyes remain anatomically normal
human eyes with a natural iris and pupil, natural sclera and normal
proportions; NEVER whiten, blank out, cloud over, glow, enlarge or
otherwise alter eyeballs, skin or anatomy — unless the story
explicitly declares that being non-human or supernatural in form.

PROPS FACE THE RIGHT WAY: every handheld or used object must be
oriented exactly as its real-world use requires. A person reading,
watching or operating something (a phone, a photograph, a paper,
any device) has its functional side — screen, front, page — facing
THEIR OWN eyes; the camera then sees whatever side the staging
geometry implies (often its back). Show the functional side to the
camera ONLY when the shot text itself stages it toward the viewer.
Never flip, mirror or reverse an object's front and back.

CARRIED STATE (persist exactly — must match the neighbouring shots of this scene): In-woo is trying to turn the same boat back from inside the wheelhouse. Su-ri-young follows with the old photograph, while her backpack, maps, coordinate note, and phone remain aboard.

PEOPLE: the SHOT TEXT alone decides whether any person is visible in this shot. IF a person appears, they must be one of: 인우 (한국인 20대 중반 남성, 짧은 검은 머리, 반듯한 얼굴형, 젊은 인상) — never anyone else, and never add a person the shot text does not show. IF only part of a person is in frame (a hand, arm, foot, back, silhouette), that body part belongs to the specific person the shot text names — its sex, age, build, skin and grooming must unmistakably match that person's profile above.

No text, captions, watermarks or annotations anywhere.

Create ONE FINAL photorealistic live-action film still of the moment below — contemporary South Korea, 2026; all people are Korean unless stated. TIME OF DAY (lock): dusk.

SHOT TEXT (authoritative, Korean): 인우의 두 손이 조타실의 나무 타륜을 꽉 쥐고 팔뚝에 힘줄이 불거진 손 클로즈업

LOCATION (lock): Inside the small fishing boat's wheelhouse at sunset, directly at the wooden helm. The shot takes place here — the attached LOCATION PHOTOGRAPH shows the exact spot.

CAMERA & FRAME (follow exactly — this is the composition authority for this still; the reference images supply identity and place, never the framing):
- CAMERA: At wheel height inside the wheelhouse, the camera completes its tilt across 인우's tensed forearms and settles on both hands clamped around the wooden wheel. The hands and forearms occupy the central two-thirds, with the wheel rim and spokes held below the nonhuman-object occupancy limit and the surrounding wheelhouse reduced to narrow edge context.
- FRAMING SCALE: close-up
- KEY BACKGROUND ELEMENTS: wooden ship's wheel (Gripped tightly by both hands) — The rim and spokes are seen obliquely from wheel height, crossing behind both forearms; used as Primary tactile anchor beneath 인우's hands, retaining enough visible rim and spokes to identify the steering action; wheelhouse interior (Visible only at the frame edges); used as Narrow peripheral context that establishes the shot inside the wheelhouse.
Compose the frame exactly as specified above — camera angle, subject scale and screen placement. Keep true physical scale between people and background elements: fixtures and distant objects occupy only the small screen area their real size and distance dictate; never enlarge a background object into a foreground presence.

REALIZE FIGURATIVE LANGUAGE AS A LIVE-ACTION SHOT: the Korean shot
text may describe characters metaphorically, figuratively or with
exaggeration. Photograph what a real movie camera would actually
record on a physical set — exaggerated or figurative impressions
become realistic staging choices (lighting, distance, angle,
wardrobe), not literal fantasy imagery.
EVERY CHARACTER IS A HUMAN BEING: unless the story explicitly
features non-human or virtual beings (as in science fiction or
fantasy), every character — however indirectly, vaguely or
figuratively the text describes them — IS a real human. When the
text gives no direct visual description of a person, IMAGINE one
and still show them as a concrete, fully-formed human being:
always render the human form to the maximum extent — build,
posture, face, hands, clothing — never reduce a person to a shape,
blob, solid silhouette or abstract mass.

EXPRESSIONS ARE ACTED, NEVER ANATOMICAL: when the text describes a
person's eyes, face or presence emotionally or figuratively (vacant,
hollow, dazed, burning, lifeless gaze and the like), it describes an
ACTOR'S PERFORMANCE captured by a real camera — realize it ONLY
through gaze direction, focus, eyelids, facial muscles, stillness
and posture. Every living person's eyes remain anatomically normal
human eyes with a natural iris and pupil, natural sclera and normal
proportions; NEVER whiten, blank out, cloud over, glow, enlarge or
otherwise alter eyeballs, skin or anatomy — unless the story
explicitly declares that being non-human or supernatural in form.

PROPS FACE THE RIGHT WAY: every handheld or used object must be
oriented exactly as its real-world use requires. A person reading,
watching or operating something (a phone, a photograph, a paper,
any device) has its functional side — screen, front, page — facing
THEIR OWN eyes; the camera then sees whatever side the staging
geometry implies (often its back). Show the functional side to the
camera ONLY when the shot text itself stages it toward the viewer.
Never flip, mirror or reverse an object's front and back.

CARRIED STATE (persist exactly — must match the neighbouring shots of this scene): In-woo is trying to turn the same boat back from inside the wheelhouse. Su-ri-young follows with the old photograph, while her backpack, maps, coordinate note, and phone remain aboard.

PEOPLE: the SHOT TEXT alone decides whether any person is visible in this shot. IF a person appears, they must be one of: 인우 (한국인 20대 중반 남성, 짧은 검은 머리, 반듯한 얼굴형, 젊은 인상) — never anyone else, and never add a person the shot text does not show. IF only part of a person is in frame (a hand, arm, foot, back, silhouette), that body part belongs to the specific person the shot text names — its sex, age, build, skin and grooming must unmistakably match that person's profile above.

No text, captions, watermarks or annotations anywhere.

Edit this photograph. Apply ONLY the corrections listed below,
each in place, changing nothing else:

CORRECTIONS:
- Adjust the back hand so it firmly grips the wooden rim of the ship's wheel.
- Change the visible sleeve to match the blue jacket from the character reference.

PRESERVE EVERYTHING ELSE EXACTLY as in the original photograph —
every person, pose, face, garment, object, set feature, lighting,
camera and framing stay identical. No text, captions or
watermarks anywhere.

Create ONE FINAL photorealistic live-action film still of the moment below — contemporary South Korea, 2026; all people are Korean unless stated. TIME OF DAY (lock): dusk.

SHOT TEXT (authoritative, Korean): 피가 흐르는 숲속에서 습격당한 남자의 허벅지에 찍힌 커다란 이빨 자국 클로즈업

LOCATION (lock): On the ground amid dense woodland undergrowth at dusk, where the injured man is being pursued. The shot takes place here — the attached LOCATION PHOTOGRAPH shows the exact spot.

CAMERA & FRAME (follow exactly — this is the composition authority for this still; the reference images supply identity and place, never the framing):
- CAMERA: From upper-thigh height on the injured side, the camera holds a tight lateral close-up while continuing the stage's dolly-in, keeping the bite mark and flowing blood in the center-right of frame. The man's limping leg crosses the shallow focal plane while the surrounding undergrowth remains only as restrained edge context, preparing the rise toward his face.
- FRAMING SCALE: close-up
- KEY BACKGROUND ELEMENTS: 우거진 수풀 (Dense enough to be pushed through); used as Soft peripheral context behind and around the moving thigh; 흐르는 피 (Actively flowing from the bite wound); used as Primary crimson detail tracing movement down the injured leg.
Compose the frame exactly as specified above — camera angle, subject scale and screen placement. Keep true physical scale between people and background elements: fixtures and distant objects occupy only the small screen area their real size and distance dictate; never enlarge a background object into a foreground presence.

REALIZE FIGURATIVE LANGUAGE AS A LIVE-ACTION SHOT: the Korean shot
text may describe characters metaphorically, figuratively or with
exaggeration. Photograph what a real movie camera would actually
record on a physical set — exaggerated or figurative impressions
become realistic staging choices (lighting, distance, angle,
wardrobe), not literal fantasy imagery.
EVERY CHARACTER IS A HUMAN BEING: unless the story explicitly
features non-human or virtual beings (as in science fiction or
fantasy), every character — however indirectly, vaguely or
figuratively the text describes them — IS a real human. When the
text gives no direct visual description of a person, IMAGINE one
and still show them as a concrete, fully-formed human being:
always render the human form to the maximum extent — build,
posture, face, hands, clothing — never reduce a person to a shape,
blob, solid silhouette or abstract mass.

EXPRESSIONS ARE ACTED, NEVER ANATOMICAL: when the text describes a
person's eyes, face or presence emotionally or figuratively (vacant,
hollow, dazed, burning, lifeless gaze and the like), it describes an
ACTOR'S PERFORMANCE captured by a real camera — realize it ONLY
through gaze direction, focus, eyelids, facial muscles, stillness
and posture. Every living person's eyes remain anatomically normal
human eyes with a natural iris and pupil, natural sclera and normal
proportions; NEVER whiten, blank out, cloud over, glow, enlarge or
otherwise alter eyeballs, skin or anatomy — unless the story
explicitly declares that being non-human or supernatural in form.

PROPS FACE THE RIGHT WAY: every handheld or used object must be
oriented exactly as its real-world use requires. A person reading,
watching or operating something (a phone, a photograph, a paper,
any device) has its functional side — screen, front, page — facing
THEIR OWN eyes; the camera then sees whatever side the staging
geometry implies (often its back). Show the functional side to the
camera ONLY when the shot text itself stages it toward the viewer.
Never flip, mirror or reverse an object's front and back.

CARRIED STATE (persist exactly — must match the neighbouring shots of this scene): The man is missing one shoe, and the large bite mark on his thigh continues to bleed.

PEOPLE: the SHOT TEXT alone decides whether any person is visible in this shot. IF a person appears, they must be one of: 습격당한 남자 (한국인 성인 남성, 짧은 검은 머리, 자연스러운 얼굴형) — never anyone else, and never add a person the shot text does not show. IF only part of a person is in frame (a hand, arm, foot, back, silhouette), that body part belongs to the specific person the shot text names — its sex, age, build, skin and grooming must unmistakably match that person's profile above.

No text, captions, watermarks or annotations anywhere.

Create ONE FINAL photorealistic live-action film still of the moment below — contemporary South Korea, 2026; all people are Korean unless stated. TIME OF DAY (lock): dusk.

SHOT TEXT (authoritative, Korean): 피가 흐르는 숲속에서 습격당한 남자의 허벅지에 찍힌 커다란 이빨 자국 클로즈업

LOCATION (lock): On the ground amid dense woodland undergrowth at dusk, where the injured man is being pursued. The shot takes place here — the attached LOCATION PHOTOGRAPH shows the exact spot.

CAMERA & FRAME (follow exactly — this is the composition authority for this still; the reference images supply identity and place, never the framing):
- CAMERA: From upper-thigh height on the injured side, the camera holds a tight lateral close-up while continuing the stage's dolly-in, keeping the bite mark and flowing blood in the center-right of frame. The man's limping leg crosses the shallow focal plane while the surrounding undergrowth remains only as restrained edge context, preparing the rise toward his face.
- FRAMING SCALE: close-up
- KEY BACKGROUND ELEMENTS: 우거진 수풀 (Dense enough to be pushed through); used as Soft peripheral context behind and around the moving thigh; 흐르는 피 (Actively flowing from the bite wound); used as Primary crimson detail tracing movement down the injured leg.
Compose the frame exactly as specified above — camera angle, subject scale and screen placement. Keep true physical scale between people and background elements: fixtures and distant objects occupy only the small screen area their real size and distance dictate; never enlarge a background object into a foreground presence.

REALIZE FIGURATIVE LANGUAGE AS A LIVE-ACTION SHOT: the Korean shot
text may describe characters metaphorically, figuratively or with
exaggeration. Photograph what a real movie camera would actually
record on a physical set — exaggerated or figurative impressions
become realistic staging choices (lighting, distance, angle,
wardrobe), not literal fantasy imagery.
EVERY CHARACTER IS A HUMAN BEING: unless the story explicitly
features non-human or virtual beings (as in science fiction or
fantasy), every character — however indirectly, vaguely or
figuratively the text describes them — IS a real human. When the
text gives no direct visual description of a person, IMAGINE one
and still show them as a concrete, fully-formed human being:
always render the human form to the maximum extent — build,
posture, face, hands, clothing — never reduce a person to a shape,
blob, solid silhouette or abstract mass.

EXPRESSIONS ARE ACTED, NEVER ANATOMICAL: when the text describes a
person's eyes, face or presence emotionally or figuratively (vacant,
hollow, dazed, burning, lifeless gaze and the like), it describes an
ACTOR'S PERFORMANCE captured by a real camera — realize it ONLY
through gaze direction, focus, eyelids, facial muscles, stillness
and posture. Every living person's eyes remain anatomically normal
human eyes with a natural iris and pupil, natural sclera and normal
proportions; NEVER whiten, blank out, cloud over, glow, enlarge or
otherwise alter eyeballs, skin or anatomy — unless the story
explicitly declares that being non-human or supernatural in form.

PROPS FACE THE RIGHT WAY: every handheld or used object must be
oriented exactly as its real-world use requires. A person reading,
watching or operating something (a phone, a photograph, a paper,
any device) has its functional side — screen, front, page — facing
THEIR OWN eyes; the camera then sees whatever side the staging
geometry implies (often its back). Show the functional side to the
camera ONLY when the shot text itself stages it toward the viewer.
Never flip, mirror or reverse an object's front and back.

CARRIED STATE (persist exactly — must match the neighbouring shots of this scene): The man is missing one shoe, and the large bite mark on his thigh continues to bleed.

PEOPLE: the SHOT TEXT alone decides whether any person is visible in this shot. IF a person appears, they must be one of: 습격당한 남자 (한국인 성인 남성, 짧은 검은 머리, 자연스러운 얼굴형) — never anyone else, and never add a person the shot text does not show. IF only part of a person is in frame (a hand, arm, foot, back, silhouette), that body part belongs to the specific person the shot text names — its sex, age, build, skin and grooming must unmistakably match that person's profile above.

No text, captions, watermarks or annotations anywhere.

VISUAL APPROACH — this take (camera, framing, composition):
Make the continuing dolly-in feel urgent and bodily, with an extremely shallow focus locking crisply onto the torn bite impression and the glossy blood while the crossing thigh carries a slight directional motion smear at its edges. Use soft, cool dusk skylight grazing laterally across the skin so the wound texture and wet crimson trail separate from the subdued, defocused undergrowth. Keep the image intimate, unstable, and viscerally immediate without losing photorealistic anatomical detail.

PRIORITY OF INSTRUCTIONS: the VISUAL APPROACH above governs camera,
framing and composition ONLY. Every other instruction in this prompt —
location, time of day, people and their identity, poses, props,
carried state, physical rules, and the no-text rule — is binding
exactly as written earlier. If the VISUAL APPROACH appears to conflict
with any of them, the earlier instruction wins.

Create ONE FINAL photorealistic live-action film still of the moment below — contemporary South Korea, 2026; all people are Korean unless stated. TIME OF DAY (lock): dusk.

SHOT TEXT (authoritative, Korean): 피가 흐르는 숲속에서 습격당한 남자의 허벅지에 찍힌 커다란 이빨 자국 클로즈업

LOCATION (lock): On the ground amid dense woodland undergrowth at dusk, where the injured man is being pursued. The shot takes place here — the attached LOCATION PHOTOGRAPH shows the exact spot.

CAMERA & FRAME (follow exactly — this is the composition authority for this still; the reference images supply identity and place, never the framing):
- CAMERA: From upper-thigh height on the injured side, the camera holds a tight lateral close-up while continuing the stage's dolly-in, keeping the bite mark and flowing blood in the center-right of frame. The man's limping leg crosses the shallow focal plane while the surrounding undergrowth remains only as restrained edge context, preparing the rise toward his face.
- FRAMING SCALE: close-up
- KEY BACKGROUND ELEMENTS: 우거진 수풀 (Dense enough to be pushed through); used as Soft peripheral context behind and around the moving thigh; 흐르는 피 (Actively flowing from the bite wound); used as Primary crimson detail tracing movement down the injured leg.
Compose the frame exactly as specified above — camera angle, subject scale and screen placement. Keep true physical scale between people and background elements: fixtures and distant objects occupy only the small screen area their real size and distance dictate; never enlarge a background object into a foreground presence.

REALIZE FIGURATIVE LANGUAGE AS A LIVE-ACTION SHOT: the Korean shot
text may describe characters metaphorically, figuratively or with
exaggeration. Photograph what a real movie camera would actually
record on a physical set — exaggerated or figurative impressions
become realistic staging choices (lighting, distance, angle,
wardrobe), not literal fantasy imagery.
EVERY CHARACTER IS A HUMAN BEING: unless the story explicitly
features non-human or virtual beings (as in science fiction or
fantasy), every character — however indirectly, vaguely or
figuratively the text describes them — IS a real human. When the
text gives no direct visual description of a person, IMAGINE one
and still show them as a concrete, fully-formed human being:
always render the human form to the maximum extent — build,
posture, face, hands, clothing — never reduce a person to a shape,
blob, solid silhouette or abstract mass.

EXPRESSIONS ARE ACTED, NEVER ANATOMICAL: when the text describes a
person's eyes, face or presence emotionally or figuratively (vacant,
hollow, dazed, burning, lifeless gaze and the like), it describes an
ACTOR'S PERFORMANCE captured by a real camera — realize it ONLY
through gaze direction, focus, eyelids, facial muscles, stillness
and posture. Every living person's eyes remain anatomically normal
human eyes with a natural iris and pupil, natural sclera and normal
proportions; NEVER whiten, blank out, cloud over, glow, enlarge or
otherwise alter eyeballs, skin or anatomy — unless the story
explicitly declares that being non-human or supernatural in form.

PROPS FACE THE RIGHT WAY: every handheld or used object must be
oriented exactly as its real-world use requires. A person reading,
watching or operating something (a phone, a photograph, a paper,
any device) has its functional side — screen, front, page — facing
THEIR OWN eyes; the camera then sees whatever side the staging
geometry implies (often its back). Show the functional side to the
camera ONLY when the shot text itself stages it toward the viewer.
Never flip, mirror or reverse an object's front and back.

CARRIED STATE (persist exactly — must match the neighbouring shots of this scene): The man is missing one shoe, and the large bite mark on his thigh continues to bleed.

PEOPLE: the SHOT TEXT alone decides whether any person is visible in this shot. IF a person appears, they must be one of: 습격당한 남자 (한국인 성인 남성, 짧은 검은 머리, 자연스러운 얼굴형) — never anyone else, and never add a person the shot text does not show. IF only part of a person is in frame (a hand, arm, foot, back, silhouette), that body part belongs to the specific person the shot text names — its sex, age, build, skin and grooming must unmistakably match that person's profile above.

No text, captions, watermarks or annotations anywhere.

VISUAL APPROACH — this take (camera, framing, composition):
Photograph the lateral close-up through a restrained layer of softly blurred undergrowth at the frame edges, creating compressed depth and the sense of the moving thigh being momentarily caught between enclosing foliage. Let the dim dusk ambience fall more evenly and naturalistically, with the wound and flowing blood emerging through controlled color contrast rather than hard highlights. Keep the dolly-in measured and ominous, preserving a clean focal hold on the center-right injury as the surrounding woodland closes into muted texture.

PRIORITY OF INSTRUCTIONS: the VISUAL APPROACH above governs camera,
framing and composition ONLY. Every other instruction in this prompt —
location, time of day, people and their identity, poses, props,
carried state, physical rules, and the no-text rule — is binding
exactly as written earlier. If the VISUAL APPROACH appears to conflict
with any of them, the earlier instruction wins.

Edit this photograph. Apply ONLY the corrections listed below,
each in place, changing nothing else:

CORRECTIONS:
- Zoom in to a tight lateral close-up on the thigh so the wound and flowing blood dominate the center-right of the frame.
- Change the semi-circular cut on the exposed skin into a large, distinct bite mark with visible individual teeth punctures.

PRESERVE EVERYTHING ELSE EXACTLY as in the original photograph —
every person, pose, face, garment, object, set feature, lighting,
camera and framing stay identical. No text, captions or
watermarks anywhere.

Create ONE FINAL photorealistic live-action film still of the moment below — contemporary South Korea, 2026; all people are Korean unless stated. TIME OF DAY (lock): night.

SHOT TEXT (authoritative, Korean): 정체불명의 아시아인 남성의 붉은 눈동자와 날카로운 이빨이 도시 뒷골목에서 습격당한 여자의 목 살점을 강하게 물어뜯어 당기는 동작 중간의 측면 클로즈업

LOCATION (lock): Beside a utility pole in a dimly lit urban back alley at night, under faint streetlight. The shot takes place here — the attached LOCATION PHOTOGRAPH shows the exact spot.

CAMERA & FRAME (follow exactly — this is the composition authority for this still; the reference images supply identity and place, never the framing):
- CAMERA: At the woman's neck level, the camera tracks perpendicular to the biting direction in a compressed side close-up, placing the attacker's red eye in the upper third and his teeth pulling her neck at center. Their overlapping profiles occupy most of the frame, while the utility pole and alley fall into narrow peripheral context without breaking the established side axis.
- FRAMING SCALE: close-up
- KEY BACKGROUND ELEMENTS: 전봇대 (The two figures are entangled beside it) — Its side faces the camera behind the two profiles; used as Narrow background anchor tying the overlapping bodies to the alley location; 가로등 (On, emitting faint light) — Its illuminated side is angled toward the alley; used as Visible background light source held out of primary focus.
Compose the frame exactly as specified above — camera angle, subject scale and screen placement. Keep true physical scale between people and background elements: fixtures and distant objects occupy only the small screen area their real size and distance dictate; never enlarge a background object into a foreground presence.

REALIZE FIGURATIVE LANGUAGE AS A LIVE-ACTION SHOT: the Korean shot
text may describe characters metaphorically, figuratively or with
exaggeration. Photograph what a real movie camera would actually
record on a physical set — exaggerated or figurative impressions
become realistic staging choices (lighting, distance, angle,
wardrobe), not literal fantasy imagery.
EVERY CHARACTER IS A HUMAN BEING: unless the story explicitly
features non-human or virtual beings (as in science fiction or
fantasy), every character — however indirectly, vaguely or
figuratively the text describes them — IS a real human. When the
text gives no direct visual description of a person, IMAGINE one
and still show them as a concrete, fully-formed human being:
always render the human form to the maximum extent — build,
posture, face, hands, clothing — never reduce a person to a shape,
blob, solid silhouette or abstract mass.

EXPRESSIONS ARE ACTED, NEVER ANATOMICAL: when the text describes a
person's eyes, face or presence emotionally or figuratively (vacant,
hollow, dazed, burning, lifeless gaze and the like), it describes an
ACTOR'S PERFORMANCE captured by a real camera — realize it ONLY
through gaze direction, focus, eyelids, facial muscles, stillness
and posture. Every living person's eyes remain anatomically normal
human eyes with a natural iris and pupil, natural sclera and normal
proportions; NEVER whiten, blank out, cloud over, glow, enlarge or
otherwise alter eyeballs, skin or anatomy — unless the story
explicitly declares that being non-human or supernatural in form.

PROPS FACE THE RIGHT WAY: every handheld or used object must be
oriented exactly as its real-world use requires. A person reading,
watching or operating something (a phone, a photograph, a paper,
any device) has its functional side — screen, front, page — facing
THEIR OWN eyes; the camera then sees whatever side the staging
geometry implies (often its back). Show the functional side to the
camera ONLY when the shot text itself stages it toward the viewer.
Never flip, mirror or reverse an object's front and back.

PEOPLE: the SHOT TEXT alone decides whether any person is visible in this shot. IF a person appears, they must be one of: 습격당한 여자 (한국인 성인 여성, 어두운색 머리카락, 자연스러운 얼굴형); 정체불명의 아시아인 남성 (포식 활성 상태) (동아시아계 성인 남성, 짧은 검은 머리, 자연스러운 남성 얼굴) — never anyone else, and never add a person the shot text does not show. IF only part of a person is in frame (a hand, arm, foot, back, silhouette), that body part belongs to the specific person the shot text names — its sex, age, build, skin and grooming must unmistakably match that person's profile above.

No text, captions, watermarks or annotations anywhere.

Create ONE FINAL photorealistic live-action film still of the moment below — contemporary South Korea, 2026; all people are Korean unless stated. TIME OF DAY (lock): night.

SHOT TEXT (authoritative, Korean): 정체불명의 아시아인 남성의 붉은 눈동자와 날카로운 이빨이 도시 뒷골목에서 습격당한 여자의 목 살점을 강하게 물어뜯어 당기는 동작 중간의 측면 클로즈업

LOCATION (lock): Beside a utility pole in a dimly lit urban back alley at night, under faint streetlight. The shot takes place here — the attached LOCATION PHOTOGRAPH shows the exact spot.

CAMERA & FRAME (follow exactly — this is the composition authority for this still; the reference images supply identity and place, never the framing):
- CAMERA: At the woman's neck level, the camera tracks perpendicular to the biting direction in a compressed side close-up, placing the attacker's red eye in the upper third and his teeth pulling her neck at center. Their overlapping profiles occupy most of the frame, while the utility pole and alley fall into narrow peripheral context without breaking the established side axis.
- FRAMING SCALE: close-up
- KEY BACKGROUND ELEMENTS: 전봇대 (The two figures are entangled beside it) — Its side faces the camera behind the two profiles; used as Narrow background anchor tying the overlapping bodies to the alley location; 가로등 (On, emitting faint light) — Its illuminated side is angled toward the alley; used as Visible background light source held out of primary focus.
Compose the frame exactly as specified above — camera angle, subject scale and screen placement. Keep true physical scale between people and background elements: fixtures and distant objects occupy only the small screen area their real size and distance dictate; never enlarge a background object into a foreground presence.

REALIZE FIGURATIVE LANGUAGE AS A LIVE-ACTION SHOT: the Korean shot
text may describe characters metaphorically, figuratively or with
exaggeration. Photograph what a real movie camera would actually
record on a physical set — exaggerated or figurative impressions
become realistic staging choices (lighting, distance, angle,
wardrobe), not literal fantasy imagery.
EVERY CHARACTER IS A HUMAN BEING: unless the story explicitly
features non-human or virtual beings (as in science fiction or
fantasy), every character — however indirectly, vaguely or
figuratively the text describes them — IS a real human. When the
text gives no direct visual description of a person, IMAGINE one
and still show them as a concrete, fully-formed human being:
always render the human form to the maximum extent — build,
posture, face, hands, clothing — never reduce a person to a shape,
blob, solid silhouette or abstract mass.

EXPRESSIONS ARE ACTED, NEVER ANATOMICAL: when the text describes a
person's eyes, face or presence emotionally or figuratively (vacant,
hollow, dazed, burning, lifeless gaze and the like), it describes an
ACTOR'S PERFORMANCE captured by a real camera — realize it ONLY
through gaze direction, focus, eyelids, facial muscles, stillness
and posture. Every living person's eyes remain anatomically normal
human eyes with a natural iris and pupil, natural sclera and normal
proportions; NEVER whiten, blank out, cloud over, glow, enlarge or
otherwise alter eyeballs, skin or anatomy — unless the story
explicitly declares that being non-human or supernatural in form.

PROPS FACE THE RIGHT WAY: every handheld or used object must be
oriented exactly as its real-world use requires. A person reading,
watching or operating something (a phone, a photograph, a paper,
any device) has its functional side — screen, front, page — facing
THEIR OWN eyes; the camera then sees whatever side the staging
geometry implies (often its back). Show the functional side to the
camera ONLY when the shot text itself stages it toward the viewer.
Never flip, mirror or reverse an object's front and back.

PEOPLE: the SHOT TEXT alone decides whether any person is visible in this shot. IF a person appears, they must be one of: 습격당한 여자 (한국인 성인 여성, 어두운색 머리카락, 자연스러운 얼굴형); 정체불명의 아시아인 남성 (포식 활성 상태) (동아시아계 성인 남성, 짧은 검은 머리, 자연스러운 남성 얼굴) — never anyone else, and never add a person the shot text does not show. IF only part of a person is in frame (a hand, arm, foot, back, silhouette), that body part belongs to the specific person the shot text names — its sex, age, build, skin and grooming must unmistakably match that person's profile above.

No text, captions, watermarks or annotations anywhere.

VISUAL APPROACH — this take (camera, framing, composition):
Use an intensely shallow focus plane that holds the attacker’s red iris, sharp teeth, and the pinched point of the woman’s neck in crisp detail while both profiles fall off rapidly toward the frame edges. Let the faint streetlight rake across the side-facing features as a narrow, hard-edged highlight, leaving the overlapping faces and the utility pole in dense low-key shadow. Keep the composition controlled and nearly still, emphasizing the horrifying precision of the bite and pull.

PRIORITY OF INSTRUCTIONS: the VISUAL APPROACH above governs camera,
framing and composition ONLY. Every other instruction in this prompt —
location, time of day, people and their identity, poses, props,
carried state, physical rules, and the no-text rule — is binding
exactly as written earlier. If the VISUAL APPROACH appears to conflict
with any of them, the earlier instruction wins.

Create ONE FINAL photorealistic live-action film still of the moment below — contemporary South Korea, 2026; all people are Korean unless stated. TIME OF DAY (lock): night.

SHOT TEXT (authoritative, Korean): 정체불명의 아시아인 남성의 붉은 눈동자와 날카로운 이빨이 도시 뒷골목에서 습격당한 여자의 목 살점을 강하게 물어뜯어 당기는 동작 중간의 측면 클로즈업

LOCATION (lock): Beside a utility pole in a dimly lit urban back alley at night, under faint streetlight. The shot takes place here — the attached LOCATION PHOTOGRAPH shows the exact spot.

CAMERA & FRAME (follow exactly — this is the composition authority for this still; the reference images supply identity and place, never the framing):
- CAMERA: At the woman's neck level, the camera tracks perpendicular to the biting direction in a compressed side close-up, placing the attacker's red eye in the upper third and his teeth pulling her neck at center. Their overlapping profiles occupy most of the frame, while the utility pole and alley fall into narrow peripheral context without breaking the established side axis.
- FRAMING SCALE: close-up
- KEY BACKGROUND ELEMENTS: 전봇대 (The two figures are entangled beside it) — Its side faces the camera behind the two profiles; used as Narrow background anchor tying the overlapping bodies to the alley location; 가로등 (On, emitting faint light) — Its illuminated side is angled toward the alley; used as Visible background light source held out of primary focus.
Compose the frame exactly as specified above — camera angle, subject scale and screen placement. Keep true physical scale between people and background elements: fixtures and distant objects occupy only the small screen area their real size and distance dictate; never enlarge a background object into a foreground presence.

REALIZE FIGURATIVE LANGUAGE AS A LIVE-ACTION SHOT: the Korean shot
text may describe characters metaphorically, figuratively or with
exaggeration. Photograph what a real movie camera would actually
record on a physical set — exaggerated or figurative impressions
become realistic staging choices (lighting, distance, angle,
wardrobe), not literal fantasy imagery.
EVERY CHARACTER IS A HUMAN BEING: unless the story explicitly
features non-human or virtual beings (as in science fiction or
fantasy), every character — however indirectly, vaguely or
figuratively the text describes them — IS a real human. When the
text gives no direct visual description of a person, IMAGINE one
and still show them as a concrete, fully-formed human being:
always render the human form to the maximum extent — build,
posture, face, hands, clothing — never reduce a person to a shape,
blob, solid silhouette or abstract mass.

EXPRESSIONS ARE ACTED, NEVER ANATOMICAL: when the text describes a
person's eyes, face or presence emotionally or figuratively (vacant,
hollow, dazed, burning, lifeless gaze and the like), it describes an
ACTOR'S PERFORMANCE captured by a real camera — realize it ONLY
through gaze direction, focus, eyelids, facial muscles, stillness
and posture. Every living person's eyes remain anatomically normal
human eyes with a natural iris and pupil, natural sclera and normal
proportions; NEVER whiten, blank out, cloud over, glow, enlarge or
otherwise alter eyeballs, skin or anatomy — unless the story
explicitly declares that being non-human or supernatural in form.

PROPS FACE THE RIGHT WAY: every handheld or used object must be
oriented exactly as its real-world use requires. A person reading,
watching or operating something (a phone, a photograph, a paper,
any device) has its functional side — screen, front, page — facing
THEIR OWN eyes; the camera then sees whatever side the staging
geometry implies (often its back). Show the functional side to the
camera ONLY when the shot text itself stages it toward the viewer.
Never flip, mirror or reverse an object's front and back.

PEOPLE: the SHOT TEXT alone decides whether any person is visible in this shot. IF a person appears, they must be one of: 습격당한 여자 (한국인 성인 여성, 어두운색 머리카락, 자연스러운 얼굴형); 정체불명의 아시아인 남성 (포식 활성 상태) (동아시아계 성인 남성, 짧은 검은 머리, 자연스러운 남성 얼굴) — never anyone else, and never add a person the shot text does not show. IF only part of a person is in frame (a hand, arm, foot, back, silhouette), that body part belongs to the specific person the shot text names — its sex, age, build, skin and grooming must unmistakably match that person's profile above.

No text, captions, watermarks or annotations anywhere.

VISUAL APPROACH — this take (camera, framing, composition):
Photograph the close-up with a deeper, more tactile focus so both overlapping profiles, the side of the utility pole, and the distant glowing streetlight remain legible across compressed layers. Preserve sharp emphasis on the attacker’s upper-third eye while allowing a restrained directional motion smear around the forceful neck-pulling action, giving the perpendicular tracking shot a sudden, bodily impact. Use the faint streetlight as soft uneven spill through the alley, favoring raw texture and claustrophobic depth over polished separation.

PRIORITY OF INSTRUCTIONS: the VISUAL APPROACH above governs camera,
framing and composition ONLY. Every other instruction in this prompt —
location, time of day, people and their identity, poses, props,
carried state, physical rules, and the no-text rule — is binding
exactly as written earlier. If the VISUAL APPROACH appears to conflict
with any of them, the earlier instruction wins.

Edit this photograph. Apply ONLY the corrections listed below,
each in place, changing nothing else:

CORRECTIONS:
- Open the man's mouth to show sharp teeth forcefully biting into and pulling the flesh of the woman's neck at the center of the frame.
- Change the color of the man's irises to red.

PRESERVE EVERYTHING ELSE EXACTLY as in the original photograph —
every person, pose, face, garment, object, set feature, lighting,
camera and framing stay identical. No text, captions or
watermarks anywhere.

Create ONE FINAL photorealistic live-action film still of the moment below — contemporary South Korea, 2026; all people are Korean unless stated. TIME OF DAY (lock): day.

SHOT TEXT (authoritative, Korean): 혜수가 형사 무리 중앙에 서서 한쪽 손을 뻗어 현장을 가리킨 채 멈춘 정면

LOCATION (lock): Inside the police-line perimeter at a daytime outdoor crime scene, among detectives directing the investigation. The shot takes place here — the attached LOCATION PHOTOGRAPH shows the exact spot.

CAMERA & FRAME (follow exactly — this is the composition authority for this still; the reference images supply identity and place, never the framing):
- CAMERA: At 혜수's eye level directly beyond the nearest detectives, the camera settles into a static medium frontal composition, with 혜수 centered and her extended hand leading diagonally toward the off-screen investigation area. The nearest detectives remain cropped at the lower side edges as restrained over-shoulder layers, preserving her command position without obscuring her gesture.
- FRAMING SCALE: medium shot
- FRAME LAYOUT: 혜수 in the middle-center of the frame, midground, points to forensic scene; forensic scene in the middle-right of the frame, background.
- KEY BACKGROUND ELEMENTS: 폴리스 라인 (Set up around the scene); used as Boundary line crossing the background and separating command personnel from recovery activity; 현장 감식 및 시신 수거 구역 (Police are processing the scene and recovering bodies); used as Background operational context beyond 혜수's directing hand.
Compose the frame exactly as specified above — camera angle, subject scale and screen placement. Keep true physical scale between people and background elements: fixtures and distant objects occupy only the small screen area their real size and distance dictate; never enlarge a background object into a foreground presence.

REALIZE FIGURATIVE LANGUAGE AS A LIVE-ACTION SHOT: the Korean shot
text may describe characters metaphorically, figuratively or with
exaggeration. Photograph what a real movie camera would actually
record on a physical set — exaggerated or figurative impressions
become realistic staging choices (lighting, distance, angle,
wardrobe), not literal fantasy imagery.
EVERY CHARACTER IS A HUMAN BEING: unless the story explicitly
features non-human or virtual beings (as in science fiction or
fantasy), every character — however indirectly, vaguely or
figuratively the text describes them — IS a real human. When the
text gives no direct visual description of a person, IMAGINE one
and still show them as a concrete, fully-formed human being:
always render the human form to the maximum extent — build,
posture, face, hands, clothing — never reduce a person to a shape,
blob, solid silhouette or abstract mass.

EXPRESSIONS ARE ACTED, NEVER ANATOMICAL: when the text describes a
person's eyes, face or presence emotionally or figuratively (vacant,
hollow, dazed, burning, lifeless gaze and the like), it describes an
ACTOR'S PERFORMANCE captured by a real camera — realize it ONLY
through gaze direction, focus, eyelids, facial muscles, stillness
and posture. Every living person's eyes remain anatomically normal
human eyes with a natural iris and pupil, natural sclera and normal
proportions; NEVER whiten, blank out, cloud over, glow, enlarge or
otherwise alter eyeballs, skin or anatomy — unless the story
explicitly declares that being non-human or supernatural in form.

PROPS FACE THE RIGHT WAY: every handheld or used object must be
oriented exactly as its real-world use requires. A person reading,
watching or operating something (a phone, a photograph, a paper,
any device) has its functional side — screen, front, page — facing
THEIR OWN eyes; the camera then sees whatever side the staging
geometry implies (often its back). Show the functional side to the
camera ONLY when the shot text itself stages it toward the viewer.
Never flip, mirror or reverse an object's front and back.

CARRIED STATE (persist exactly — must match the neighbouring shots of this scene): The attack site remains secured with police tape, with the recovered victim’s body still part of the active crime scene.

PEOPLE: the SHOT TEXT alone decides whether any person is visible in this shot. IF a person appears, they must be one of: 혜수 (한국인 40대 초반 여성, 단정한 검은 머리, 성숙한 타원형 얼굴) — never anyone else, and never add a person the shot text does not show. IF only part of a person is in frame (a hand, arm, foot, back, silhouette), that body part belongs to the specific person the shot text names — its sex, age, build, skin and grooming must unmistakably match that person's profile above.

No text, captions, watermarks or annotations anywhere.

Create ONE FINAL photorealistic live-action film still of the moment below — contemporary South Korea, 2026; all people are Korean unless stated. TIME OF DAY (lock): day.

SHOT TEXT (authoritative, Korean): 혜수가 형사 무리 중앙에 서서 한쪽 손을 뻗어 현장을 가리킨 채 멈춘 정면

LOCATION (lock): Inside the police-line perimeter at a daytime outdoor crime scene, among detectives directing the investigation. The shot takes place here — the attached LOCATION PHOTOGRAPH shows the exact spot.

CAMERA & FRAME (follow exactly — this is the composition authority for this still; the reference images supply identity and place, never the framing):
- CAMERA: At 혜수's eye level directly beyond the nearest detectives, the camera settles into a static medium frontal composition, with 혜수 centered and her extended hand leading diagonally toward the off-screen investigation area. The nearest detectives remain cropped at the lower side edges as restrained over-shoulder layers, preserving her command position without obscuring her gesture.
- FRAMING SCALE: medium shot
- FRAME LAYOUT: 혜수 in the middle-center of the frame, midground, points to forensic scene; forensic scene in the middle-right of the frame, background.
- KEY BACKGROUND ELEMENTS: 폴리스 라인 (Set up around the scene); used as Boundary line crossing the background and separating command personnel from recovery activity; 현장 감식 및 시신 수거 구역 (Police are processing the scene and recovering bodies); used as Background operational context beyond 혜수's directing hand.
Compose the frame exactly as specified above — camera angle, subject scale and screen placement. Keep true physical scale between people and background elements: fixtures and distant objects occupy only the small screen area their real size and distance dictate; never enlarge a background object into a foreground presence.

REALIZE FIGURATIVE LANGUAGE AS A LIVE-ACTION SHOT: the Korean shot
text may describe characters metaphorically, figuratively or with
exaggeration. Photograph what a real movie camera would actually
record on a physical set — exaggerated or figurative impressions
become realistic staging choices (lighting, distance, angle,
wardrobe), not literal fantasy imagery.
EVERY CHARACTER IS A HUMAN BEING: unless the story explicitly
features non-human or virtual beings (as in science fiction or
fantasy), every character — however indirectly, vaguely or
figuratively the text describes them — IS a real human. When the
text gives no direct visual description of a person, IMAGINE one
and still show them as a concrete, fully-formed human being:
always render the human form to the maximum extent — build,
posture, face, hands, clothing — never reduce a person to a shape,
blob, solid silhouette or abstract mass.

EXPRESSIONS ARE ACTED, NEVER ANATOMICAL: when the text describes a
person's eyes, face or presence emotionally or figuratively (vacant,
hollow, dazed, burning, lifeless gaze and the like), it describes an
ACTOR'S PERFORMANCE captured by a real camera — realize it ONLY
through gaze direction, focus, eyelids, facial muscles, stillness
and posture. Every living person's eyes remain anatomically normal
human eyes with a natural iris and pupil, natural sclera and normal
proportions; NEVER whiten, blank out, cloud over, glow, enlarge or
otherwise alter eyeballs, skin or anatomy — unless the story
explicitly declares that being non-human or supernatural in form.

PROPS FACE THE RIGHT WAY: every handheld or used object must be
oriented exactly as its real-world use requires. A person reading,
watching or operating something (a phone, a photograph, a paper,
any device) has its functional side — screen, front, page — facing
THEIR OWN eyes; the camera then sees whatever side the staging
geometry implies (often its back). Show the functional side to the
camera ONLY when the shot text itself stages it toward the viewer.
Never flip, mirror or reverse an object's front and back.

CARRIED STATE (persist exactly — must match the neighbouring shots of this scene): The attack site remains secured with police tape, with the recovered victim’s body still part of the active crime scene.

PEOPLE: the SHOT TEXT alone decides whether any person is visible in this shot. IF a person appears, they must be one of: 혜수 (한국인 40대 초반 여성, 단정한 검은 머리, 성숙한 타원형 얼굴) — never anyone else, and never add a person the shot text does not show. IF only part of a person is in frame (a hand, arm, foot, back, silhouette), that body part belongs to the specific person the shot text names — its sex, age, build, skin and grooming must unmistakably match that person's profile above.

No text, captions, watermarks or annotations anywhere.

VISUAL APPROACH — this take (camera, framing, composition):
Use a restrained, slightly compressed lens feel with disciplined deep focus, keeping 혜수 crisply dominant while the police-line boundary and recovery activity remain clearly legible at their natural distance. Shape the daytime light evenly and realistically, emphasizing the firm diagonal of her extended hand and the orderly depth from the cropped edge layers to the active background.

PRIORITY OF INSTRUCTIONS: the VISUAL APPROACH above governs camera,
framing and composition ONLY. Every other instruction in this prompt —
location, time of day, people and their identity, poses, props,
carried state, physical rules, and the no-text rule — is binding
exactly as written earlier. If the VISUAL APPROACH appears to conflict
with any of them, the earlier instruction wins.

Create ONE FINAL photorealistic live-action film still of the moment below — contemporary South Korea, 2026; all people are Korean unless stated. TIME OF DAY (lock): day.

SHOT TEXT (authoritative, Korean): 혜수가 형사 무리 중앙에 서서 한쪽 손을 뻗어 현장을 가리킨 채 멈춘 정면

LOCATION (lock): Inside the police-line perimeter at a daytime outdoor crime scene, among detectives directing the investigation. The shot takes place here — the attached LOCATION PHOTOGRAPH shows the exact spot.

CAMERA & FRAME (follow exactly — this is the composition authority for this still; the reference images supply identity and place, never the framing):
- CAMERA: At 혜수's eye level directly beyond the nearest detectives, the camera settles into a static medium frontal composition, with 혜수 centered and her extended hand leading diagonally toward the off-screen investigation area. The nearest detectives remain cropped at the lower side edges as restrained over-shoulder layers, preserving her command position without obscuring her gesture.
- FRAMING SCALE: medium shot
- FRAME LAYOUT: 혜수 in the middle-center of the frame, midground, points to forensic scene; forensic scene in the middle-right of the frame, background.
- KEY BACKGROUND ELEMENTS: 폴리스 라인 (Set up around the scene); used as Boundary line crossing the background and separating command personnel from recovery activity; 현장 감식 및 시신 수거 구역 (Police are processing the scene and recovering bodies); used as Background operational context beyond 혜수's directing hand.
Compose the frame exactly as specified above — camera angle, subject scale and screen placement. Keep true physical scale between people and background elements: fixtures and distant objects occupy only the small screen area their real size and distance dictate; never enlarge a background object into a foreground presence.

REALIZE FIGURATIVE LANGUAGE AS A LIVE-ACTION SHOT: the Korean shot
text may describe characters metaphorically, figuratively or with
exaggeration. Photograph what a real movie camera would actually
record on a physical set — exaggerated or figurative impressions
become realistic staging choices (lighting, distance, angle,
wardrobe), not literal fantasy imagery.
EVERY CHARACTER IS A HUMAN BEING: unless the story explicitly
features non-human or virtual beings (as in science fiction or
fantasy), every character — however indirectly, vaguely or
figuratively the text describes them — IS a real human. When the
text gives no direct visual description of a person, IMAGINE one
and still show them as a concrete, fully-formed human being:
always render the human form to the maximum extent — build,
posture, face, hands, clothing — never reduce a person to a shape,
blob, solid silhouette or abstract mass.

EXPRESSIONS ARE ACTED, NEVER ANATOMICAL: when the text describes a
person's eyes, face or presence emotionally or figuratively (vacant,
hollow, dazed, burning, lifeless gaze and the like), it describes an
ACTOR'S PERFORMANCE captured by a real camera — realize it ONLY
through gaze direction, focus, eyelids, facial muscles, stillness
and posture. Every living person's eyes remain anatomically normal
human eyes with a natural iris and pupil, natural sclera and normal
proportions; NEVER whiten, blank out, cloud over, glow, enlarge or
otherwise alter eyeballs, skin or anatomy — unless the story
explicitly declares that being non-human or supernatural in form.

PROPS FACE THE RIGHT WAY: every handheld or used object must be
oriented exactly as its real-world use requires. A person reading,
watching or operating something (a phone, a photograph, a paper,
any device) has its functional side — screen, front, page — facing
THEIR OWN eyes; the camera then sees whatever side the staging
geometry implies (often its back). Show the functional side to the
camera ONLY when the shot text itself stages it toward the viewer.
Never flip, mirror or reverse an object's front and back.

CARRIED STATE (persist exactly — must match the neighbouring shots of this scene): The attack site remains secured with police tape, with the recovered victim’s body still part of the active crime scene.

PEOPLE: the SHOT TEXT alone decides whether any person is visible in this shot. IF a person appears, they must be one of: 혜수 (한국인 40대 초반 여성, 단정한 검은 머리, 성숙한 타원형 얼굴) — never anyone else, and never add a person the shot text does not show. IF only part of a person is in frame (a hand, arm, foot, back, silhouette), that body part belongs to the specific person the shot text names — its sex, age, build, skin and grooming must unmistakably match that person's profile above.

No text, captions, watermarks or annotations anywhere.

VISUAL APPROACH — this take (camera, framing, composition):
Use a natural-perspective lens feel with selective focus centered on 혜수’s face and extended hand, allowing the cropped nearest layers and distant recovery activity to fall progressively softer while remaining recognizable. Let directional daytime light give her face and gesture subtle definition, creating a tense visual pause against the subdued operational depth behind her.

PRIORITY OF INSTRUCTIONS: the VISUAL APPROACH above governs camera,
framing and composition ONLY. Every other instruction in this prompt —
location, time of day, people and their identity, poses, props,
carried state, physical rules, and the no-text rule — is binding
exactly as written earlier. If the VISUAL APPROACH appears to conflict
with any of them, the earlier instruction wins.

Edit this photograph. Apply ONLY the corrections listed below,
each in place, changing nothing else:

CORRECTIONS:
- Remove all text from the police tape, leaving it as a solid, blank color band.
- Correct the anatomy of the pointing hand to clearly show only the index finger extended, with the other fingers naturally curled into the palm.

PRESERVE EVERYTHING ELSE EXACTLY as in the original photograph —
every person, pose, face, garment, object, set feature, lighting,
camera and framing stay identical. No text, captions or
watermarks anywhere.

Create ONE FINAL photorealistic live-action film still of the moment below — contemporary South Korea, 2026; all people are Korean unless stated. TIME OF DAY (lock): day.

SHOT TEXT (authoritative, Korean): 이어폰을 낀 수리영이 해안가 도로에서 자전거 안장에 앉아 한쪽 페달을 밟아 내리는 중간 찰나의 전신, 머리카락과 옷자락이 이동 반대 방향으로 흩날리고 있는 상태

LOCATION (lock): On the open coastal road beside the sea in daylight, along the lane where she rides her bicycle. The shot takes place here — the attached LOCATION PHOTOGRAPH shows the exact spot.

CAMERA & FRAME (follow exactly — this is the composition authority for this still; the reference images supply identity and place, never the framing):
- CAMERA: At waist height directly beside the bicycle, the camera tracks parallel in a full-body side view timed to one pedal reaching its downward stroke. 수리영 and the bicycle occupy the middle band moving toward frame right, while her hair and clothing trail toward frame left and the coastal road remains visible as scale context.
- FRAMING SCALE: wide shot
- FRAME LAYOUT: 수리영 in the middle-center of the frame, midground, moves toward road ahead at frame right; coastal road ahead in the middle-right of the frame, background.
- KEY BACKGROUND ELEMENTS: 자전거 (In forward motion) — Its full side faces the camera as it travels parallel to the lens; used as Side-profile movement platform and full-body scale reference; 해안가 도로 (Daytime road); used as Linear background that establishes the cyclist's direction of travel; 바다 (Visible from the coastal route); used as Location context beside the road.
Compose the frame exactly as specified above — camera angle, subject scale and screen placement. Keep true physical scale between people and background elements: fixtures and distant objects occupy only the small screen area their real size and distance dictate; never enlarge a background object into a foreground presence.

REALIZE FIGURATIVE LANGUAGE AS A LIVE-ACTION SHOT: the Korean shot
text may describe characters metaphorically, figuratively or with
exaggeration. Photograph what a real movie camera would actually
record on a physical set — exaggerated or figurative impressions
become realistic staging choices (lighting, distance, angle,
wardrobe), not literal fantasy imagery.
EVERY CHARACTER IS A HUMAN BEING: unless the story explicitly
features non-human or virtual beings (as in science fiction or
fantasy), every character — however indirectly, vaguely or
figuratively the text describes them — IS a real human. When the
text gives no direct visual description of a person, IMAGINE one
and still show them as a concrete, fully-formed human being:
always render the human form to the maximum extent — build,
posture, face, hands, clothing — never reduce a person to a shape,
blob, solid silhouette or abstract mass.

EXPRESSIONS ARE ACTED, NEVER ANATOMICAL: when the text describes a
person's eyes, face or presence emotionally or figuratively (vacant,
hollow, dazed, burning, lifeless gaze and the like), it describes an
ACTOR'S PERFORMANCE captured by a real camera — realize it ONLY
through gaze direction, focus, eyelids, facial muscles, stillness
and posture. Every living person's eyes remain anatomically normal
human eyes with a natural iris and pupil, natural sclera and normal
proportions; NEVER whiten, blank out, cloud over, glow, enlarge or
otherwise alter eyeballs, skin or anatomy — unless the story
explicitly declares that being non-human or supernatural in form.

PROPS FACE THE RIGHT WAY: every handheld or used object must be
oriented exactly as its real-world use requires. A person reading,
watching or operating something (a phone, a photograph, a paper,
any device) has its functional side — screen, front, page — facing
THEIR OWN eyes; the camera then sees whatever side the staging
geometry implies (often its back). Show the functional side to the
camera ONLY when the shot text itself stages it toward the viewer.
Never flip, mirror or reverse an object's front and back.

CARRIED STATE (persist exactly — must match the neighbouring shots of this scene): Su-ri-young is riding her bicycle with her earphones still in place.

PEOPLE: the SHOT TEXT alone decides whether any person is visible in this shot. IF a person appears, they must be one of: 수리영 (한국인 20대 초반 여성, 긴 검은 생머리, 젊은 타원형 얼굴) — never anyone else, and never add a person the shot text does not show. IF only part of a person is in frame (a hand, arm, foot, back, silhouette), that body part belongs to the specific person the shot text names — its sex, age, build, skin and grooming must unmistakably match that person's profile above.

No text, captions, watermarks or annotations anywhere.

Create ONE FINAL photorealistic live-action film still of the moment below — contemporary South Korea, 2026; all people are Korean unless stated. TIME OF DAY (lock): day.

SHOT TEXT (authoritative, Korean): 이어폰을 낀 수리영이 해안가 도로에서 자전거 안장에 앉아 한쪽 페달을 밟아 내리는 중간 찰나의 전신, 머리카락과 옷자락이 이동 반대 방향으로 흩날리고 있는 상태

LOCATION (lock): On the open coastal road beside the sea in daylight, along the lane where she rides her bicycle. The shot takes place here — the attached LOCATION PHOTOGRAPH shows the exact spot.

CAMERA & FRAME (follow exactly — this is the composition authority for this still; the reference images supply identity and place, never the framing):
- CAMERA: At waist height directly beside the bicycle, the camera tracks parallel in a full-body side view timed to one pedal reaching its downward stroke. 수리영 and the bicycle occupy the middle band moving toward frame right, while her hair and clothing trail toward frame left and the coastal road remains visible as scale context.
- FRAMING SCALE: wide shot
- FRAME LAYOUT: 수리영 in the middle-center of the frame, midground, moves toward road ahead at frame right; coastal road ahead in the middle-right of the frame, background.
- KEY BACKGROUND ELEMENTS: 자전거 (In forward motion) — Its full side faces the camera as it travels parallel to the lens; used as Side-profile movement platform and full-body scale reference; 해안가 도로 (Daytime road); used as Linear background that establishes the cyclist's direction of travel; 바다 (Visible from the coastal route); used as Location context beside the road.
Compose the frame exactly as specified above — camera angle, subject scale and screen placement. Keep true physical scale between people and background elements: fixtures and distant objects occupy only the small screen area their real size and distance dictate; never enlarge a background object into a foreground presence.

REALIZE FIGURATIVE LANGUAGE AS A LIVE-ACTION SHOT: the Korean shot
text may describe characters metaphorically, figuratively or with
exaggeration. Photograph what a real movie camera would actually
record on a physical set — exaggerated or figurative impressions
become realistic staging choices (lighting, distance, angle,
wardrobe), not literal fantasy imagery.
EVERY CHARACTER IS A HUMAN BEING: unless the story explicitly
features non-human or virtual beings (as in science fiction or
fantasy), every character — however indirectly, vaguely or
figuratively the text describes them — IS a real human. When the
text gives no direct visual description of a person, IMAGINE one
and still show them as a concrete, fully-formed human being:
always render the human form to the maximum extent — build,
posture, face, hands, clothing — never reduce a person to a shape,
blob, solid silhouette or abstract mass.

EXPRESSIONS ARE ACTED, NEVER ANATOMICAL: when the text describes a
person's eyes, face or presence emotionally or figuratively (vacant,
hollow, dazed, burning, lifeless gaze and the like), it describes an
ACTOR'S PERFORMANCE captured by a real camera — realize it ONLY
through gaze direction, focus, eyelids, facial muscles, stillness
and posture. Every living person's eyes remain anatomically normal
human eyes with a natural iris and pupil, natural sclera and normal
proportions; NEVER whiten, blank out, cloud over, glow, enlarge or
otherwise alter eyeballs, skin or anatomy — unless the story
explicitly declares that being non-human or supernatural in form.

PROPS FACE THE RIGHT WAY: every handheld or used object must be
oriented exactly as its real-world use requires. A person reading,
watching or operating something (a phone, a photograph, a paper,
any device) has its functional side — screen, front, page — facing
THEIR OWN eyes; the camera then sees whatever side the staging
geometry implies (often its back). Show the functional side to the
camera ONLY when the shot text itself stages it toward the viewer.
Never flip, mirror or reverse an object's front and back.

CARRIED STATE (persist exactly — must match the neighbouring shots of this scene): Su-ri-young is riding her bicycle with her earphones still in place.

PEOPLE: the SHOT TEXT alone decides whether any person is visible in this shot. IF a person appears, they must be one of: 수리영 (한국인 20대 초반 여성, 긴 검은 생머리, 젊은 타원형 얼굴) — never anyone else, and never add a person the shot text does not show. IF only part of a person is in frame (a hand, arm, foot, back, silhouette), that body part belongs to the specific person the shot text names — its sex, age, build, skin and grooming must unmistakably match that person's profile above.

No text, captions, watermarks or annotations anywhere.

VISUAL APPROACH — this take (camera, framing, composition):
Use a fluid parallel tracking treatment at the prescribed waist-height wide side view, holding Su-ri-young and the bicycle sharply legible while the coastal road and sea carry restrained horizontal motion blur. Favor a natural perspective with layered depth, clear separation of her full figure and both wheels, and daylight shaped gently across her face and body. Let the trailing hair and clothing edges provide the strongest visible counterflow to the movement toward frame right.

PRIORITY OF INSTRUCTIONS: the VISUAL APPROACH above governs camera,
framing and composition ONLY. Every other instruction in this prompt —
location, time of day, people and their identity, poses, props,
carried state, physical rules, and the no-text rule — is binding
exactly as written earlier. If the VISUAL APPROACH appears to conflict
with any of them, the earlier instruction wins.

Create ONE FINAL photorealistic live-action film still of the moment below — contemporary South Korea, 2026; all people are Korean unless stated. TIME OF DAY (lock): day.

SHOT TEXT (authoritative, Korean): 이어폰을 낀 수리영이 해안가 도로에서 자전거 안장에 앉아 한쪽 페달을 밟아 내리는 중간 찰나의 전신, 머리카락과 옷자락이 이동 반대 방향으로 흩날리고 있는 상태

LOCATION (lock): On the open coastal road beside the sea in daylight, along the lane where she rides her bicycle. The shot takes place here — the attached LOCATION PHOTOGRAPH shows the exact spot.

CAMERA & FRAME (follow exactly — this is the composition authority for this still; the reference images supply identity and place, never the framing):
- CAMERA: At waist height directly beside the bicycle, the camera tracks parallel in a full-body side view timed to one pedal reaching its downward stroke. 수리영 and the bicycle occupy the middle band moving toward frame right, while her hair and clothing trail toward frame left and the coastal road remains visible as scale context.
- FRAMING SCALE: wide shot
- FRAME LAYOUT: 수리영 in the middle-center of the frame, midground, moves toward road ahead at frame right; coastal road ahead in the middle-right of the frame, background.
- KEY BACKGROUND ELEMENTS: 자전거 (In forward motion) — Its full side faces the camera as it travels parallel to the lens; used as Side-profile movement platform and full-body scale reference; 해안가 도로 (Daytime road); used as Linear background that establishes the cyclist's direction of travel; 바다 (Visible from the coastal route); used as Location context beside the road.
Compose the frame exactly as specified above — camera angle, subject scale and screen placement. Keep true physical scale between people and background elements: fixtures and distant objects occupy only the small screen area their real size and distance dictate; never enlarge a background object into a foreground presence.

REALIZE FIGURATIVE LANGUAGE AS A LIVE-ACTION SHOT: the Korean shot
text may describe characters metaphorically, figuratively or with
exaggeration. Photograph what a real movie camera would actually
record on a physical set — exaggerated or figurative impressions
become realistic staging choices (lighting, distance, angle,
wardrobe), not literal fantasy imagery.
EVERY CHARACTER IS A HUMAN BEING: unless the story explicitly
features non-human or virtual beings (as in science fiction or
fantasy), every character — however indirectly, vaguely or
figuratively the text describes them — IS a real human. When the
text gives no direct visual description of a person, IMAGINE one
and still show them as a concrete, fully-formed human being:
always render the human form to the maximum extent — build,
posture, face, hands, clothing — never reduce a person to a shape,
blob, solid silhouette or abstract mass.

EXPRESSIONS ARE ACTED, NEVER ANATOMICAL: when the text describes a
person's eyes, face or presence emotionally or figuratively (vacant,
hollow, dazed, burning, lifeless gaze and the like), it describes an
ACTOR'S PERFORMANCE captured by a real camera — realize it ONLY
through gaze direction, focus, eyelids, facial muscles, stillness
and posture. Every living person's eyes remain anatomically normal
human eyes with a natural iris and pupil, natural sclera and normal
proportions; NEVER whiten, blank out, cloud over, glow, enlarge or
otherwise alter eyeballs, skin or anatomy — unless the story
explicitly declares that being non-human or supernatural in form.

PROPS FACE THE RIGHT WAY: every handheld or used object must be
oriented exactly as its real-world use requires. A person reading,
watching or operating something (a phone, a photograph, a paper,
any device) has its functional side — screen, front, page — facing
THEIR OWN eyes; the camera then sees whatever side the staging
geometry implies (often its back). Show the functional side to the
camera ONLY when the shot text itself stages it toward the viewer.
Never flip, mirror or reverse an object's front and back.

CARRIED STATE (persist exactly — must match the neighbouring shots of this scene): Su-ri-young is riding her bicycle with her earphones still in place.

PEOPLE: the SHOT TEXT alone decides whether any person is visible in this shot. IF a person appears, they must be one of: 수리영 (한국인 20대 초반 여성, 긴 검은 생머리, 젊은 타원형 얼굴) — never anyone else, and never add a person the shot text does not show. IF only part of a person is in frame (a hand, arm, foot, back, silhouette), that body part belongs to the specific person the shot text names — its sex, age, build, skin and grooming must unmistakably match that person's profile above.

No text, captions, watermarks or annotations anywhere.

VISUAL APPROACH — this take (camera, framing, composition):
Photograph the prescribed waist-height wide side view as a crisp, suspended instant with deep focus, preserving fine detail from Su-ri-young and the full bicycle through the coastal road and visible sea. Build a clean lateral composition from the bicycle’s geometry and the road’s receding line, with the downward pedal stroke as the visual fulcrum and generous environmental context around her. Use firm natural daylight and restrained contrast to define her profile, wind-swept hair, and trailing clothing without romantic haze or pronounced motion smear.

PRIORITY OF INSTRUCTIONS: the VISUAL APPROACH above governs camera,
framing and composition ONLY. Every other instruction in this prompt —
location, time of day, people and their identity, poses, props,
carried state, physical rules, and the no-text rule — is binding
exactly as written earlier. If the VISUAL APPROACH appears to conflict
with any of them, the earlier instruction wins.

Edit this photograph. Apply ONLY the corrections listed below,
each in place, changing nothing else:

CORRECTIONS:
- Redirect the hair so it trails backwards toward frame left, following the physics of her forward motion.
- Add the visible portions of her left leg and foot positioned on the opposite pedal.

PRESERVE EVERYTHING ELSE EXACTLY as in the original photograph —
every person, pose, face, garment, object, set feature, lighting,
camera and framing stay identical. No text, captions or
watermarks anywhere.

Create ONE FINAL photorealistic live-action film still of the moment below — contemporary South Korea, 2026; all people are Korean unless stated. TIME OF DAY (lock): day.

SHOT TEXT (authoritative, Korean): 수리영이 매서운 눈빛으로 로트와일러를 똑바로 응시하는 정면 클로즈업

LOCATION (lock): In the dog-handling area of an animal shelter, directly in front of the agitated Rottweiler. The shot takes place here — the attached LOCATION PHOTOGRAPH shows the exact spot.

CAMERA & FRAME (follow exactly — this is the composition authority for this still; the reference images supply identity and place, never the framing):
- CAMERA: At 수리영's eye level at the closest point of the dolly-in, the camera holds a centered frontal close-up on her unwavering stare, with her face occupying most of the frame. The back of the 로트와일러's lowered head remains as a restrained, soft foreground edge below her eyeline, maintaining the confrontation for the coming arc around the dog.
- FRAMING SCALE: close-up
- FRAME LAYOUT: 수리영 in the middle-center of the frame, midground, looks toward Rottweiler below the lens line; 로트와일러 in the lower-center of the frame, foreground, looks toward Suriyoung at frame center.
- KEY BACKGROUND ELEMENTS: 보호원 울타리와 우리 (Containing the shelter animals) — Only partial front-facing enclosure lines remain visible behind 수리영; used as Soft peripheral location context behind the confrontation.
Compose the frame exactly as specified above — camera angle, subject scale and screen placement. Keep true physical scale between people and background elements: fixtures and distant objects occupy only the small screen area their real size and distance dictate; never enlarge a background object into a foreground presence.

REALIZE FIGURATIVE LANGUAGE AS A LIVE-ACTION SHOT: the Korean shot
text may describe characters metaphorically, figuratively or with
exaggeration. Photograph what a real movie camera would actually
record on a physical set — exaggerated or figurative impressions
become realistic staging choices (lighting, distance, angle,
wardrobe), not literal fantasy imagery.
EVERY CHARACTER IS A HUMAN BEING: unless the story explicitly
features non-human or virtual beings (as in science fiction or
fantasy), every character — however indirectly, vaguely or
figuratively the text describes them — IS a real human. When the
text gives no direct visual description of a person, IMAGINE one
and still show them as a concrete, fully-formed human being:
always render the human form to the maximum extent — build,
posture, face, hands, clothing — never reduce a person to a shape,
blob, solid silhouette or abstract mass.

EXPRESSIONS ARE ACTED, NEVER ANATOMICAL: when the text describes a
person's eyes, face or presence emotionally or figuratively (vacant,
hollow, dazed, burning, lifeless gaze and the like), it describes an
ACTOR'S PERFORMANCE captured by a real camera — realize it ONLY
through gaze direction, focus, eyelids, facial muscles, stillness
and posture. Every living person's eyes remain anatomically normal
human eyes with a natural iris and pupil, natural sclera and normal
proportions; NEVER whiten, blank out, cloud over, glow, enlarge or
otherwise alter eyeballs, skin or anatomy — unless the story
explicitly declares that being non-human or supernatural in form.

PROPS FACE THE RIGHT WAY: every handheld or used object must be
oriented exactly as its real-world use requires. A person reading,
watching or operating something (a phone, a photograph, a paper,
any device) has its functional side — screen, front, page — facing
THEIR OWN eyes; the camera then sees whatever side the staging
geometry implies (often its back). Show the functional side to the
camera ONLY when the shot text itself stages it toward the viewer.
Never flip, mirror or reverse an object's front and back.

PEOPLE: the SHOT TEXT alone decides whether any person is visible in this shot. IF a person appears, they must be one of: 수리영 (한국인 20대 초반 여성, 긴 검은 생머리, 젊은 타원형 얼굴) — never anyone else, and never add a person the shot text does not show. IF only part of a person is in frame (a hand, arm, foot, back, silhouette), that body part belongs to the specific person the shot text names — its sex, age, build, skin and grooming must unmistakably match that person's profile above.

No text, captions, watermarks or annotations anywhere.

Create ONE FINAL photorealistic live-action film still of the moment below — contemporary South Korea, 2026; all people are Korean unless stated. TIME OF DAY (lock): day.

SHOT TEXT (authoritative, Korean): 수리영이 매서운 눈빛으로 로트와일러를 똑바로 응시하는 정면 클로즈업

LOCATION (lock): In the dog-handling area of an animal shelter, directly in front of the agitated Rottweiler. The shot takes place here — the attached LOCATION PHOTOGRAPH shows the exact spot.

CAMERA & FRAME (follow exactly — this is the composition authority for this still; the reference images supply identity and place, never the framing):
- CAMERA: At 수리영's eye level at the closest point of the dolly-in, the camera holds a centered frontal close-up on her unwavering stare, with her face occupying most of the frame. The back of the 로트와일러's lowered head remains as a restrained, soft foreground edge below her eyeline, maintaining the confrontation for the coming arc around the dog.
- FRAMING SCALE: close-up
- FRAME LAYOUT: 수리영 in the middle-center of the frame, midground, looks toward Rottweiler below the lens line; 로트와일러 in the lower-center of the frame, foreground, looks toward Suriyoung at frame center.
- KEY BACKGROUND ELEMENTS: 보호원 울타리와 우리 (Containing the shelter animals) — Only partial front-facing enclosure lines remain visible behind 수리영; used as Soft peripheral location context behind the confrontation.
Compose the frame exactly as specified above — camera angle, subject scale and screen placement. Keep true physical scale between people and background elements: fixtures and distant objects occupy only the small screen area their real size and distance dictate; never enlarge a background object into a foreground presence.

REALIZE FIGURATIVE LANGUAGE AS A LIVE-ACTION SHOT: the Korean shot
text may describe characters metaphorically, figuratively or with
exaggeration. Photograph what a real movie camera would actually
record on a physical set — exaggerated or figurative impressions
become realistic staging choices (lighting, distance, angle,
wardrobe), not literal fantasy imagery.
EVERY CHARACTER IS A HUMAN BEING: unless the story explicitly
features non-human or virtual beings (as in science fiction or
fantasy), every character — however indirectly, vaguely or
figuratively the text describes them — IS a real human. When the
text gives no direct visual description of a person, IMAGINE one
and still show them as a concrete, fully-formed human being:
always render the human form to the maximum extent — build,
posture, face, hands, clothing — never reduce a person to a shape,
blob, solid silhouette or abstract mass.

EXPRESSIONS ARE ACTED, NEVER ANATOMICAL: when the text describes a
person's eyes, face or presence emotionally or figuratively (vacant,
hollow, dazed, burning, lifeless gaze and the like), it describes an
ACTOR'S PERFORMANCE captured by a real camera — realize it ONLY
through gaze direction, focus, eyelids, facial muscles, stillness
and posture. Every living person's eyes remain anatomically normal
human eyes with a natural iris and pupil, natural sclera and normal
proportions; NEVER whiten, blank out, cloud over, glow, enlarge or
otherwise alter eyeballs, skin or anatomy — unless the story
explicitly declares that being non-human or supernatural in form.

PROPS FACE THE RIGHT WAY: every handheld or used object must be
oriented exactly as its real-world use requires. A person reading,
watching or operating something (a phone, a photograph, a paper,
any device) has its functional side — screen, front, page — facing
THEIR OWN eyes; the camera then sees whatever side the staging
geometry implies (often its back). Show the functional side to the
camera ONLY when the shot text itself stages it toward the viewer.
Never flip, mirror or reverse an object's front and back.

PEOPLE: the SHOT TEXT alone decides whether any person is visible in this shot. IF a person appears, they must be one of: 수리영 (한국인 20대 초반 여성, 긴 검은 생머리, 젊은 타원형 얼굴) — never anyone else, and never add a person the shot text does not show. IF only part of a person is in frame (a hand, arm, foot, back, silhouette), that body part belongs to the specific person the shot text names — its sex, age, build, skin and grooming must unmistakably match that person's profile above.

No text, captions, watermarks or annotations anywhere.

VISUAL APPROACH — this take (camera, framing, composition):
Use a compressed long-lens feel and very shallow depth of field, isolating her eyes and minute facial tension while the lowered head below and the enclosure lines behind dissolve into restrained softness. Broad, diffused daytime light should model her face gently and evenly, preserving natural skin texture; let the locked symmetry and her absolute stillness carry the menace.

PRIORITY OF INSTRUCTIONS: the VISUAL APPROACH above governs camera,
framing and composition ONLY. Every other instruction in this prompt —
location, time of day, people and their identity, poses, props,
carried state, physical rules, and the no-text rule — is binding
exactly as written earlier. If the VISUAL APPROACH appears to conflict
with any of them, the earlier instruction wins.

Create ONE FINAL photorealistic live-action film still of the moment below — contemporary South Korea, 2026; all people are Korean unless stated. TIME OF DAY (lock): day.

SHOT TEXT (authoritative, Korean): 수리영이 매서운 눈빛으로 로트와일러를 똑바로 응시하는 정면 클로즈업

LOCATION (lock): In the dog-handling area of an animal shelter, directly in front of the agitated Rottweiler. The shot takes place here — the attached LOCATION PHOTOGRAPH shows the exact spot.

CAMERA & FRAME (follow exactly — this is the composition authority for this still; the reference images supply identity and place, never the framing):
- CAMERA: At 수리영's eye level at the closest point of the dolly-in, the camera holds a centered frontal close-up on her unwavering stare, with her face occupying most of the frame. The back of the 로트와일러's lowered head remains as a restrained, soft foreground edge below her eyeline, maintaining the confrontation for the coming arc around the dog.
- FRAMING SCALE: close-up
- FRAME LAYOUT: 수리영 in the middle-center of the frame, midground, looks toward Rottweiler below the lens line; 로트와일러 in the lower-center of the frame, foreground, looks toward Suriyoung at frame center.
- KEY BACKGROUND ELEMENTS: 보호원 울타리와 우리 (Containing the shelter animals) — Only partial front-facing enclosure lines remain visible behind 수리영; used as Soft peripheral location context behind the confrontation.
Compose the frame exactly as specified above — camera angle, subject scale and screen placement. Keep true physical scale between people and background elements: fixtures and distant objects occupy only the small screen area their real size and distance dictate; never enlarge a background object into a foreground presence.

REALIZE FIGURATIVE LANGUAGE AS A LIVE-ACTION SHOT: the Korean shot
text may describe characters metaphorically, figuratively or with
exaggeration. Photograph what a real movie camera would actually
record on a physical set — exaggerated or figurative impressions
become realistic staging choices (lighting, distance, angle,
wardrobe), not literal fantasy imagery.
EVERY CHARACTER IS A HUMAN BEING: unless the story explicitly
features non-human or virtual beings (as in science fiction or
fantasy), every character — however indirectly, vaguely or
figuratively the text describes them — IS a real human. When the
text gives no direct visual description of a person, IMAGINE one
and still show them as a concrete, fully-formed human being:
always render the human form to the maximum extent — build,
posture, face, hands, clothing — never reduce a person to a shape,
blob, solid silhouette or abstract mass.

EXPRESSIONS ARE ACTED, NEVER ANATOMICAL: when the text describes a
person's eyes, face or presence emotionally or figuratively (vacant,
hollow, dazed, burning, lifeless gaze and the like), it describes an
ACTOR'S PERFORMANCE captured by a real camera — realize it ONLY
through gaze direction, focus, eyelids, facial muscles, stillness
and posture. Every living person's eyes remain anatomically normal
human eyes with a natural iris and pupil, natural sclera and normal
proportions; NEVER whiten, blank out, cloud over, glow, enlarge or
otherwise alter eyeballs, skin or anatomy — unless the story
explicitly declares that being non-human or supernatural in form.

PROPS FACE THE RIGHT WAY: every handheld or used object must be
oriented exactly as its real-world use requires. A person reading,
watching or operating something (a phone, a photograph, a paper,
any device) has its functional side — screen, front, page — facing
THEIR OWN eyes; the camera then sees whatever side the staging
geometry implies (often its back). Show the functional side to the
camera ONLY when the shot text itself stages it toward the viewer.
Never flip, mirror or reverse an object's front and back.

PEOPLE: the SHOT TEXT alone decides whether any person is visible in this shot. IF a person appears, they must be one of: 수리영 (한국인 20대 초반 여성, 긴 검은 생머리, 젊은 타원형 얼굴) — never anyone else, and never add a person the shot text does not show. IF only part of a person is in frame (a hand, arm, foot, back, silhouette), that body part belongs to the specific person the shot text names — its sex, age, build, skin and grooming must unmistakably match that person's profile above.

No text, captions, watermarks or annotations anywhere.

VISUAL APPROACH — this take (camera, framing, composition):
Use a close, natural-perspective lens feel that makes the physical distance of the confrontation immediate without distorting her face, while retaining the prescribed centered frontal close-up and soft lower foreground edge. Shape the daytime light more directionally across her features, with controlled contrast defining the eyelids, cheekbones, and set jaw; keep the peripheral enclosure geometry slightly more legible so its rigid lines intensify the unwavering stare.

PRIORITY OF INSTRUCTIONS: the VISUAL APPROACH above governs camera,
framing and composition ONLY. Every other instruction in this prompt —
location, time of day, people and their identity, poses, props,
carried state, physical rules, and the no-text rule — is binding
exactly as written earlier. If the VISUAL APPROACH appears to conflict
with any of them, the earlier instruction wins.

Edit this photograph. Apply ONLY the corrections listed below,
each in place, changing nothing else:

CORRECTIONS:
- Angle Suriyoung's eyes downward so her gaze is locked on the dog below the lens line.

PRESERVE EVERYTHING ELSE EXACTLY as in the original photograph —
every person, pose, face, garment, object, set feature, lighting,
camera and framing stay identical. No text, captions or
watermarks anywhere.

Create ONE FINAL photorealistic live-action film still of the moment below — contemporary South Korea, 2026; all people are Korean unless stated. TIME OF DAY (lock): night.

SHOT TEXT (authoritative, Korean): 어둠이 깔린 해안도로에서 수리영이 상체를 숙인 채 한쪽 무릎을 아래로 굽혀 자전거 페달을 밟는 중간 자세, 머리카락과 옷자락이 달리는 방향 반대편으로 흩날리는 전신

LOCATION (lock): On the dark coastal road at night, along the route leading back toward her home. The shot takes place here — the attached LOCATION PHOTOGRAPH shows the exact spot.

CAMERA & FRAME (follow exactly — this is the composition authority for this still; the reference images supply identity and place, never the framing):
- CAMERA: At hip height beside the bicycle, the camera travels parallel as 수리영 rides screen-left to screen-right, lowering slightly to hold her bent knee and downward pedal stroke while preserving her entire figure. Her torso and bicycle form a taut diagonal across the center, with the road continuing ahead on the right and her trailing hair and clothing describing the opposite direction of travel.
- FRAMING SCALE: wide shot
- FRAME LAYOUT: 수리영 riding the bicycle in the middle-center of the frame, midground, moves toward screen-right road ahead; coastal road ahead in the middle-right of the frame, background.
- KEY BACKGROUND ELEMENTS: 해안도로 (In use as 수리영 rides home); used as Parallel travel line beneath and ahead of the rider; 자전거 (Moving rapidly under pedal power) — Its full side faces the camera as it travels toward screen right; used as Primary vehicle held in clean side profile with the rider.
Compose the frame exactly as specified above — camera angle, subject scale and screen placement. Keep true physical scale between people and background elements: fixtures and distant objects occupy only the small screen area their real size and distance dictate; never enlarge a background object into a foreground presence.

REALIZE FIGURATIVE LANGUAGE AS A LIVE-ACTION SHOT: the Korean shot
text may describe characters metaphorically, figuratively or with
exaggeration. Photograph what a real movie camera would actually
record on a physical set — exaggerated or figurative impressions
become realistic staging choices (lighting, distance, angle,
wardrobe), not literal fantasy imagery.
EVERY CHARACTER IS A HUMAN BEING: unless the story explicitly
features non-human or virtual beings (as in science fiction or
fantasy), every character — however indirectly, vaguely or
figuratively the text describes them — IS a real human. When the
text gives no direct visual description of a person, IMAGINE one
and still show them as a concrete, fully-formed human being:
always render the human form to the maximum extent — build,
posture, face, hands, clothing — never reduce a person to a shape,
blob, solid silhouette or abstract mass.

EXPRESSIONS ARE ACTED, NEVER ANATOMICAL: when the text describes a
person's eyes, face or presence emotionally or figuratively (vacant,
hollow, dazed, burning, lifeless gaze and the like), it describes an
ACTOR'S PERFORMANCE captured by a real camera — realize it ONLY
through gaze direction, focus, eyelids, facial muscles, stillness
and posture. Every living person's eyes remain anatomically normal
human eyes with a natural iris and pupil, natural sclera and normal
proportions; NEVER whiten, blank out, cloud over, glow, enlarge or
otherwise alter eyeballs, skin or anatomy — unless the story
explicitly declares that being non-human or supernatural in form.

PROPS FACE THE RIGHT WAY: every handheld or used object must be
oriented exactly as its real-world use requires. A person reading,
watching or operating something (a phone, a photograph, a paper,
any device) has its functional side — screen, front, page — facing
THEIR OWN eyes; the camera then sees whatever side the staging
geometry implies (often its back). Show the functional side to the
camera ONLY when the shot text itself stages it toward the viewer.
Never flip, mirror or reverse an object's front and back.

CARRIED STATE (persist exactly — must match the neighbouring shots of this scene): Su-ri-young continues home on the same bicycle, with her phone still in her possession.

PEOPLE: the SHOT TEXT alone decides whether any person is visible in this shot. IF a person appears, they must be one of: 수리영 (한국인 20대 초반 여성, 긴 검은 생머리, 젊은 타원형 얼굴) — never anyone else, and never add a person the shot text does not show. IF only part of a person is in frame (a hand, arm, foot, back, silhouette), that body part belongs to the specific person the shot text names — its sex, age, build, skin and grooming must unmistakably match that person's profile above.

No text, captions, watermarks or annotations anywhere.

Create ONE FINAL photorealistic live-action film still of the moment below — contemporary South Korea, 2026; all people are Korean unless stated. TIME OF DAY (lock): night.

SHOT TEXT (authoritative, Korean): 어둠이 깔린 해안도로에서 수리영이 상체를 숙인 채 한쪽 무릎을 아래로 굽혀 자전거 페달을 밟는 중간 자세, 머리카락과 옷자락이 달리는 방향 반대편으로 흩날리는 전신

LOCATION (lock): On the dark coastal road at night, along the route leading back toward her home. The shot takes place here — the attached LOCATION PHOTOGRAPH shows the exact spot.

CAMERA & FRAME (follow exactly — this is the composition authority for this still; the reference images supply identity and place, never the framing):
- CAMERA: At hip height beside the bicycle, the camera travels parallel as 수리영 rides screen-left to screen-right, lowering slightly to hold her bent knee and downward pedal stroke while preserving her entire figure. Her torso and bicycle form a taut diagonal across the center, with the road continuing ahead on the right and her trailing hair and clothing describing the opposite direction of travel.
- FRAMING SCALE: wide shot
- FRAME LAYOUT: 수리영 riding the bicycle in the middle-center of the frame, midground, moves toward screen-right road ahead; coastal road ahead in the middle-right of the frame, background.
- KEY BACKGROUND ELEMENTS: 해안도로 (In use as 수리영 rides home); used as Parallel travel line beneath and ahead of the rider; 자전거 (Moving rapidly under pedal power) — Its full side faces the camera as it travels toward screen right; used as Primary vehicle held in clean side profile with the rider.
Compose the frame exactly as specified above — camera angle, subject scale and screen placement. Keep true physical scale between people and background elements: fixtures and distant objects occupy only the small screen area their real size and distance dictate; never enlarge a background object into a foreground presence.

REALIZE FIGURATIVE LANGUAGE AS A LIVE-ACTION SHOT: the Korean shot
text may describe characters metaphorically, figuratively or with
exaggeration. Photograph what a real movie camera would actually
record on a physical set — exaggerated or figurative impressions
become realistic staging choices (lighting, distance, angle,
wardrobe), not literal fantasy imagery.
EVERY CHARACTER IS A HUMAN BEING: unless the story explicitly
features non-human or virtual beings (as in science fiction or
fantasy), every character — however indirectly, vaguely or
figuratively the text describes them — IS a real human. When the
text gives no direct visual description of a person, IMAGINE one
and still show them as a concrete, fully-formed human being:
always render the human form to the maximum extent — build,
posture, face, hands, clothing — never reduce a person to a shape,
blob, solid silhouette or abstract mass.

EXPRESSIONS ARE ACTED, NEVER ANATOMICAL: when the text describes a
person's eyes, face or presence emotionally or figuratively (vacant,
hollow, dazed, burning, lifeless gaze and the like), it describes an
ACTOR'S PERFORMANCE captured by a real camera — realize it ONLY
through gaze direction, focus, eyelids, facial muscles, stillness
and posture. Every living person's eyes remain anatomically normal
human eyes with a natural iris and pupil, natural sclera and normal
proportions; NEVER whiten, blank out, cloud over, glow, enlarge or
otherwise alter eyeballs, skin or anatomy — unless the story
explicitly declares that being non-human or supernatural in form.

PROPS FACE THE RIGHT WAY: every handheld or used object must be
oriented exactly as its real-world use requires. A person reading,
watching or operating something (a phone, a photograph, a paper,
any device) has its functional side — screen, front, page — facing
THEIR OWN eyes; the camera then sees whatever side the staging
geometry implies (often its back). Show the functional side to the
camera ONLY when the shot text itself stages it toward the viewer.
Never flip, mirror or reverse an object's front and back.

CARRIED STATE (persist exactly — must match the neighbouring shots of this scene): Su-ri-young continues home on the same bicycle, with her phone still in her possession.

PEOPLE: the SHOT TEXT alone decides whether any person is visible in this shot. IF a person appears, they must be one of: 수리영 (한국인 20대 초반 여성, 긴 검은 생머리, 젊은 타원형 얼굴) — never anyone else, and never add a person the shot text does not show. IF only part of a person is in frame (a hand, arm, foot, back, silhouette), that body part belongs to the specific person the shot text names — its sex, age, build, skin and grooming must unmistakably match that person's profile above.

No text, captions, watermarks or annotations anywhere.

VISUAL APPROACH — this take (camera, framing, composition):
Use a broad, immersive lens feel from the locked hip-height parallel tracking position, keeping Su-ri-young and the bicycle’s full side profile sharply legible while the road surface and distant coast smear subtly into horizontal motion. Let the receding road on the right pull the eye forward, with her lowered knee, pitched torso, streaming hair, and clothing creating a tense counterflow across the frame. Shape her face and body with restrained, directional night illumination, preserving natural detail against the dark surroundings.

PRIORITY OF INSTRUCTIONS: the VISUAL APPROACH above governs camera,
framing and composition ONLY. Every other instruction in this prompt —
location, time of day, people and their identity, poses, props,
carried state, physical rules, and the no-text rule — is binding
exactly as written earlier. If the VISUAL APPROACH appears to conflict
with any of them, the earlier instruction wins.

Create ONE FINAL photorealistic live-action film still of the moment below — contemporary South Korea, 2026; all people are Korean unless stated. TIME OF DAY (lock): night.

SHOT TEXT (authoritative, Korean): 어둠이 깔린 해안도로에서 수리영이 상체를 숙인 채 한쪽 무릎을 아래로 굽혀 자전거 페달을 밟는 중간 자세, 머리카락과 옷자락이 달리는 방향 반대편으로 흩날리는 전신

LOCATION (lock): On the dark coastal road at night, along the route leading back toward her home. The shot takes place here — the attached LOCATION PHOTOGRAPH shows the exact spot.

CAMERA & FRAME (follow exactly — this is the composition authority for this still; the reference images supply identity and place, never the framing):
- CAMERA: At hip height beside the bicycle, the camera travels parallel as 수리영 rides screen-left to screen-right, lowering slightly to hold her bent knee and downward pedal stroke while preserving her entire figure. Her torso and bicycle form a taut diagonal across the center, with the road continuing ahead on the right and her trailing hair and clothing describing the opposite direction of travel.
- FRAMING SCALE: wide shot
- FRAME LAYOUT: 수리영 riding the bicycle in the middle-center of the frame, midground, moves toward screen-right road ahead; coastal road ahead in the middle-right of the frame, background.
- KEY BACKGROUND ELEMENTS: 해안도로 (In use as 수리영 rides home); used as Parallel travel line beneath and ahead of the rider; 자전거 (Moving rapidly under pedal power) — Its full side faces the camera as it travels toward screen right; used as Primary vehicle held in clean side profile with the rider.
Compose the frame exactly as specified above — camera angle, subject scale and screen placement. Keep true physical scale between people and background elements: fixtures and distant objects occupy only the small screen area their real size and distance dictate; never enlarge a background object into a foreground presence.

REALIZE FIGURATIVE LANGUAGE AS A LIVE-ACTION SHOT: the Korean shot
text may describe characters metaphorically, figuratively or with
exaggeration. Photograph what a real movie camera would actually
record on a physical set — exaggerated or figurative impressions
become realistic staging choices (lighting, distance, angle,
wardrobe), not literal fantasy imagery.
EVERY CHARACTER IS A HUMAN BEING: unless the story explicitly
features non-human or virtual beings (as in science fiction or
fantasy), every character — however indirectly, vaguely or
figuratively the text describes them — IS a real human. When the
text gives no direct visual description of a person, IMAGINE one
and still show them as a concrete, fully-formed human being:
always render the human form to the maximum extent — build,
posture, face, hands, clothing — never reduce a person to a shape,
blob, solid silhouette or abstract mass.

EXPRESSIONS ARE ACTED, NEVER ANATOMICAL: when the text describes a
person's eyes, face or presence emotionally or figuratively (vacant,
hollow, dazed, burning, lifeless gaze and the like), it describes an
ACTOR'S PERFORMANCE captured by a real camera — realize it ONLY
through gaze direction, focus, eyelids, facial muscles, stillness
and posture. Every living person's eyes remain anatomically normal
human eyes with a natural iris and pupil, natural sclera and normal
proportions; NEVER whiten, blank out, cloud over, glow, enlarge or
otherwise alter eyeballs, skin or anatomy — unless the story
explicitly declares that being non-human or supernatural in form.

PROPS FACE THE RIGHT WAY: every handheld or used object must be
oriented exactly as its real-world use requires. A person reading,
watching or operating something (a phone, a photograph, a paper,
any device) has its functional side — screen, front, page — facing
THEIR OWN eyes; the camera then sees whatever side the staging
geometry implies (often its back). Show the functional side to the
camera ONLY when the shot text itself stages it toward the viewer.
Never flip, mirror or reverse an object's front and back.

CARRIED STATE (persist exactly — must match the neighbouring shots of this scene): Su-ri-young continues home on the same bicycle, with her phone still in her possession.

PEOPLE: the SHOT TEXT alone decides whether any person is visible in this shot. IF a person appears, they must be one of: 수리영 (한국인 20대 초반 여성, 긴 검은 생머리, 젊은 타원형 얼굴) — never anyone else, and never add a person the shot text does not show. IF only part of a person is in frame (a hand, arm, foot, back, silhouette), that body part belongs to the specific person the shot text names — its sex, age, build, skin and grooming must unmistakably match that person's profile above.

No text, captions, watermarks or annotations anywhere.

VISUAL APPROACH — this take (camera, framing, composition):
Use a compressed lens feel within the prescribed wide framing, flattening the coastal road behind Su-ri-young and isolating her centered diagonal posture against the night. Freeze the downward pedal stroke, bent knee, trailing hair, and lifted clothing with crisp physical clarity, allowing only the slightest wheel movement to suggest speed. Keep the illumination soft and low-key, with controlled tonal separation across her full figure and bicycle while the road ahead remains clearly readable at middle-right.

PRIORITY OF INSTRUCTIONS: the VISUAL APPROACH above governs camera,
framing and composition ONLY. Every other instruction in this prompt —
location, time of day, people and their identity, poses, props,
carried state, physical rules, and the no-text rule — is binding
exactly as written earlier. If the VISUAL APPROACH appears to conflict
with any of them, the earlier instruction wins.

Edit this photograph. Apply ONLY the corrections listed below,
each in place, changing nothing else:

CORRECTIONS:
- Change the pants to blue denim jeans to match the character reference.
- Add the green cap from the reference image to the character's head.
- Reposition the right foot so it rests naturally on the right pedal of the bicycle.

PRESERVE EVERYTHING ELSE EXACTLY as in the original photograph —
every person, pose, face, garment, object, set feature, lighting,
camera and framing stay identical. No text, captions or
watermarks anywhere.

Create ONE FINAL photorealistic live-action film still of the moment below — contemporary South Korea, 2026; all people are Korean unless stated. TIME OF DAY (lock): night.

SHOT TEXT (authoritative, Korean): 밤바다를 배경으로 버스정류장 벤치에 허리를 굽히고 멍하니 앉아 있는 수리영의 전신

LOCATION (lock): On a bench at a coastal bus stop at night, facing the dark sea. The shot takes place here — the attached LOCATION PHOTOGRAPH shows the exact spot.

CAMERA & FRAME (follow exactly — this is the composition authority for this still; the reference images supply identity and place, never the framing):
- CAMERA: From across the bench at standing eye level, hold a static wide composition with 수리영's full seated figure centered against the night sea behind her. Her bent torso and lowered head interrupt the otherwise broad horizontal separation between bench and sea, making her isolation legible before the planned forward move.
- FRAMING SCALE: wide shot
- KEY BACKGROUND ELEMENTS: bus-stop bench (Occupied by 수리영) — Its front edge faces the camera, with 수리영 seated near the center; used as Supports 수리영's full figure and establishes her solitary seated position; sea (Visible at night); used as Forms the broad background field behind 수리영.
Compose the frame exactly as specified above — camera angle, subject scale and screen placement. Keep true physical scale between people and background elements: fixtures and distant objects occupy only the small screen area their real size and distance dictate; never enlarge a background object into a foreground presence.

REALIZE FIGURATIVE LANGUAGE AS A LIVE-ACTION SHOT: the Korean shot
text may describe characters metaphorically, figuratively or with
exaggeration. Photograph what a real movie camera would actually
record on a physical set — exaggerated or figurative impressions
become realistic staging choices (lighting, distance, angle,
wardrobe), not literal fantasy imagery.
EVERY CHARACTER IS A HUMAN BEING: unless the story explicitly
features non-human or virtual beings (as in science fiction or
fantasy), every character — however indirectly, vaguely or
figuratively the text describes them — IS a real human. When the
text gives no direct visual description of a person, IMAGINE one
and still show them as a concrete, fully-formed human being:
always render the human form to the maximum extent — build,
posture, face, hands, clothing — never reduce a person to a shape,
blob, solid silhouette or abstract mass.

EXPRESSIONS ARE ACTED, NEVER ANATOMICAL: when the text describes a
person's eyes, face or presence emotionally or figuratively (vacant,
hollow, dazed, burning, lifeless gaze and the like), it describes an
ACTOR'S PERFORMANCE captured by a real camera — realize it ONLY
through gaze direction, focus, eyelids, facial muscles, stillness
and posture. Every living person's eyes remain anatomically normal
human eyes with a natural iris and pupil, natural sclera and normal
proportions; NEVER whiten, blank out, cloud over, glow, enlarge or
otherwise alter eyeballs, skin or anatomy — unless the story
explicitly declares that being non-human or supernatural in form.

PROPS FACE THE RIGHT WAY: every handheld or used object must be
oriented exactly as its real-world use requires. A person reading,
watching or operating something (a phone, a photograph, a paper,
any device) has its functional side — screen, front, page — facing
THEIR OWN eyes; the camera then sees whatever side the staging
geometry implies (often its back). Show the functional side to the
camera ONLY when the shot text itself stages it toward the viewer.
Never flip, mirror or reverse an object's front and back.

CARRIED STATE (persist exactly — must match the neighbouring shots of this scene): Su-ri-young still has her phone, but Min-suk remains unreachable and absent.

PEOPLE: the SHOT TEXT alone decides whether any person is visible in this shot. IF a person appears, they must be one of: 수리영 (한국인 20대 초반 여성, 긴 검은 생머리, 젊은 타원형 얼굴) — never anyone else, and never add a person the shot text does not show. IF only part of a person is in frame (a hand, arm, foot, back, silhouette), that body part belongs to the specific person the shot text names — its sex, age, build, skin and grooming must unmistakably match that person's profile above.

No text, captions, watermarks or annotations anywhere.

Create ONE FINAL photorealistic live-action film still of the moment below — contemporary South Korea, 2026; all people are Korean unless stated. TIME OF DAY (lock): night.

SHOT TEXT (authoritative, Korean): 밤바다를 배경으로 버스정류장 벤치에 허리를 굽히고 멍하니 앉아 있는 수리영의 전신

LOCATION (lock): On a bench at a coastal bus stop at night, facing the dark sea. The shot takes place here — the attached LOCATION PHOTOGRAPH shows the exact spot.

CAMERA & FRAME (follow exactly — this is the composition authority for this still; the reference images supply identity and place, never the framing):
- CAMERA: From across the bench at standing eye level, hold a static wide composition with 수리영's full seated figure centered against the night sea behind her. Her bent torso and lowered head interrupt the otherwise broad horizontal separation between bench and sea, making her isolation legible before the planned forward move.
- FRAMING SCALE: wide shot
- KEY BACKGROUND ELEMENTS: bus-stop bench (Occupied by 수리영) — Its front edge faces the camera, with 수리영 seated near the center; used as Supports 수리영's full figure and establishes her solitary seated position; sea (Visible at night); used as Forms the broad background field behind 수리영.
Compose the frame exactly as specified above — camera angle, subject scale and screen placement. Keep true physical scale between people and background elements: fixtures and distant objects occupy only the small screen area their real size and distance dictate; never enlarge a background object into a foreground presence.

REALIZE FIGURATIVE LANGUAGE AS A LIVE-ACTION SHOT: the Korean shot
text may describe characters metaphorically, figuratively or with
exaggeration. Photograph what a real movie camera would actually
record on a physical set — exaggerated or figurative impressions
become realistic staging choices (lighting, distance, angle,
wardrobe), not literal fantasy imagery.
EVERY CHARACTER IS A HUMAN BEING: unless the story explicitly
features non-human or virtual beings (as in science fiction or
fantasy), every character — however indirectly, vaguely or
figuratively the text describes them — IS a real human. When the
text gives no direct visual description of a person, IMAGINE one
and still show them as a concrete, fully-formed human being:
always render the human form to the maximum extent — build,
posture, face, hands, clothing — never reduce a person to a shape,
blob, solid silhouette or abstract mass.

EXPRESSIONS ARE ACTED, NEVER ANATOMICAL: when the text describes a
person's eyes, face or presence emotionally or figuratively (vacant,
hollow, dazed, burning, lifeless gaze and the like), it describes an
ACTOR'S PERFORMANCE captured by a real camera — realize it ONLY
through gaze direction, focus, eyelids, facial muscles, stillness
and posture. Every living person's eyes remain anatomically normal
human eyes with a natural iris and pupil, natural sclera and normal
proportions; NEVER whiten, blank out, cloud over, glow, enlarge or
otherwise alter eyeballs, skin or anatomy — unless the story
explicitly declares that being non-human or supernatural in form.

PROPS FACE THE RIGHT WAY: every handheld or used object must be
oriented exactly as its real-world use requires. A person reading,
watching or operating something (a phone, a photograph, a paper,
any device) has its functional side — screen, front, page — facing
THEIR OWN eyes; the camera then sees whatever side the staging
geometry implies (often its back). Show the functional side to the
camera ONLY when the shot text itself stages it toward the viewer.
Never flip, mirror or reverse an object's front and back.

CARRIED STATE (persist exactly — must match the neighbouring shots of this scene): Su-ri-young still has her phone, but Min-suk remains unreachable and absent.

PEOPLE: the SHOT TEXT alone decides whether any person is visible in this shot. IF a person appears, they must be one of: 수리영 (한국인 20대 초반 여성, 긴 검은 생머리, 젊은 타원형 얼굴) — never anyone else, and never add a person the shot text does not show. IF only part of a person is in frame (a hand, arm, foot, back, silhouette), that body part belongs to the specific person the shot text names — its sex, age, build, skin and grooming must unmistakably match that person's profile above.

No text, captions, watermarks or annotations anywhere.

VISUAL APPROACH — this take (camera, framing, composition):
Use a restrained frontal tableau with a natural wide-angle feel, keeping the full seated figure precisely centered and the bench front edge strongly horizontal. Hold clear depth through the bench and dark sea so the broad empty background remains legible, with subdued cool night illumination, gentle facial modeling, and deep surrounding blacks that preserve realistic detail in her bowed posture.

PRIORITY OF INSTRUCTIONS: the VISUAL APPROACH above governs camera,
framing and composition ONLY. Every other instruction in this prompt —
location, time of day, people and their identity, poses, props,
carried state, physical rules, and the no-text rule — is binding
exactly as written earlier. If the VISUAL APPROACH appears to conflict
with any of them, the earlier instruction wins.

Create ONE FINAL photorealistic live-action film still of the moment below — contemporary South Korea, 2026; all people are Korean unless stated. TIME OF DAY (lock): night.

SHOT TEXT (authoritative, Korean): 밤바다를 배경으로 버스정류장 벤치에 허리를 굽히고 멍하니 앉아 있는 수리영의 전신

LOCATION (lock): On a bench at a coastal bus stop at night, facing the dark sea. The shot takes place here — the attached LOCATION PHOTOGRAPH shows the exact spot.

CAMERA & FRAME (follow exactly — this is the composition authority for this still; the reference images supply identity and place, never the framing):
- CAMERA: From across the bench at standing eye level, hold a static wide composition with 수리영's full seated figure centered against the night sea behind her. Her bent torso and lowered head interrupt the otherwise broad horizontal separation between bench and sea, making her isolation legible before the planned forward move.
- FRAMING SCALE: wide shot
- KEY BACKGROUND ELEMENTS: bus-stop bench (Occupied by 수리영) — Its front edge faces the camera, with 수리영 seated near the center; used as Supports 수리영's full figure and establishes her solitary seated position; sea (Visible at night); used as Forms the broad background field behind 수리영.
Compose the frame exactly as specified above — camera angle, subject scale and screen placement. Keep true physical scale between people and background elements: fixtures and distant objects occupy only the small screen area their real size and distance dictate; never enlarge a background object into a foreground presence.

REALIZE FIGURATIVE LANGUAGE AS A LIVE-ACTION SHOT: the Korean shot
text may describe characters metaphorically, figuratively or with
exaggeration. Photograph what a real movie camera would actually
record on a physical set — exaggerated or figurative impressions
become realistic staging choices (lighting, distance, angle,
wardrobe), not literal fantasy imagery.
EVERY CHARACTER IS A HUMAN BEING: unless the story explicitly
features non-human or virtual beings (as in science fiction or
fantasy), every character — however indirectly, vaguely or
figuratively the text describes them — IS a real human. When the
text gives no direct visual description of a person, IMAGINE one
and still show them as a concrete, fully-formed human being:
always render the human form to the maximum extent — build,
posture, face, hands, clothing — never reduce a person to a shape,
blob, solid silhouette or abstract mass.

EXPRESSIONS ARE ACTED, NEVER ANATOMICAL: when the text describes a
person's eyes, face or presence emotionally or figuratively (vacant,
hollow, dazed, burning, lifeless gaze and the like), it describes an
ACTOR'S PERFORMANCE captured by a real camera — realize it ONLY
through gaze direction, focus, eyelids, facial muscles, stillness
and posture. Every living person's eyes remain anatomically normal
human eyes with a natural iris and pupil, natural sclera and normal
proportions; NEVER whiten, blank out, cloud over, glow, enlarge or
otherwise alter eyeballs, skin or anatomy — unless the story
explicitly declares that being non-human or supernatural in form.

PROPS FACE THE RIGHT WAY: every handheld or used object must be
oriented exactly as its real-world use requires. A person reading,
watching or operating something (a phone, a photograph, a paper,
any device) has its functional side — screen, front, page — facing
THEIR OWN eyes; the camera then sees whatever side the staging
geometry implies (often its back). Show the functional side to the
camera ONLY when the shot text itself stages it toward the viewer.
Never flip, mirror or reverse an object's front and back.

CARRIED STATE (persist exactly — must match the neighbouring shots of this scene): Su-ri-young still has her phone, but Min-suk remains unreachable and absent.

PEOPLE: the SHOT TEXT alone decides whether any person is visible in this shot. IF a person appears, they must be one of: 수리영 (한국인 20대 초반 여성, 긴 검은 생머리, 젊은 타원형 얼굴) — never anyone else, and never add a person the shot text does not show. IF only part of a person is in frame (a hand, arm, foot, back, silhouette), that body part belongs to the specific person the shot text names — its sex, age, build, skin and grooming must unmistakably match that person's profile above.

No text, captions, watermarks or annotations anywhere.

VISUAL APPROACH — this take (camera, framing, composition):
Give the wide shot a subtly compressed lens feel, flattening the visual distance between the centered seated figure, bench, and night sea while retaining her complete body in frame. Use selective focus to keep her posture, face, hands, and clothing crisp as the sea falls softly away, with low-key lateral night light tracing her hair and bent shoulders and a restrained falloff across the surrounding frame.

PRIORITY OF INSTRUCTIONS: the VISUAL APPROACH above governs camera,
framing and composition ONLY. Every other instruction in this prompt —
location, time of day, people and their identity, poses, props,
carried state, physical rules, and the no-text rule — is binding
exactly as written earlier. If the VISUAL APPROACH appears to conflict
with any of them, the earlier instruction wins.

Edit this photograph. Apply ONLY the corrections listed below,
each in place, changing nothing else:

CORRECTIONS:
- Add the green baseball cap from the reference image onto the character's head.

PRESERVE EVERYTHING ELSE EXACTLY as in the original photograph —
every person, pose, face, garment, object, set feature, lighting,
camera and framing stay identical. No text, captions or
watermarks anywhere.

Create ONE FINAL photorealistic live-action film still of the moment below — contemporary South Korea, 2026; all people are Korean unless stated. TIME OF DAY (lock): late morning.

SHOT TEXT (authoritative, Korean): 어두운 PC방 안, 수리영이 켜진 모니터 화면을 응시하며 앉아 있는 상체

LOCATION (lock): At a computer booth inside a dark internet café, illuminated mainly by the active monitor. The shot takes place here — the attached LOCATION PHOTOGRAPH shows the exact spot.

CAMERA & FRAME (follow exactly — this is the composition authority for this still; the reference images supply identity and place, never the framing):
- CAMERA: Begin at seated eye level in front of and slightly to one side of Suriyoung, showing her upper body on one half of the medium frame and the active monitor on the other. Start the prescribed slow dolly toward the desk without yet lowering, using her three-quarter posture and fixed eyeline to bind her to the coordinate search.
- FRAMING SCALE: medium shot
- KEY BACKGROUND ELEMENTS: PC방 모니터 (On and displaying the location search) — The active front face angles toward Suriyoung and remains partly readable to the camera as a coordinate-location display; used as Shares the frame with Suriyoung as the immediate focus of her search; PC방 책상 (In use) — Its front edge angles across the lower frame toward the monitor; used as Creates the working plane that the following downward camera move will approach.
Compose the frame exactly as specified above — camera angle, subject scale and screen placement. Keep true physical scale between people and background elements: fixtures and distant objects occupy only the small screen area their real size and distance dictate; never enlarge a background object into a foreground presence.

REALIZE FIGURATIVE LANGUAGE AS A LIVE-ACTION SHOT: the Korean shot
text may describe characters metaphorically, figuratively or with
exaggeration. Photograph what a real movie camera would actually
record on a physical set — exaggerated or figurative impressions
become realistic staging choices (lighting, distance, angle,
wardrobe), not literal fantasy imagery.
EVERY CHARACTER IS A HUMAN BEING: unless the story explicitly
features non-human or virtual beings (as in science fiction or
fantasy), every character — however indirectly, vaguely or
figuratively the text describes them — IS a real human. When the
text gives no direct visual description of a person, IMAGINE one
and still show them as a concrete, fully-formed human being:
always render the human form to the maximum extent — build,
posture, face, hands, clothing — never reduce a person to a shape,
blob, solid silhouette or abstract mass.

EXPRESSIONS ARE ACTED, NEVER ANATOMICAL: when the text describes a
person's eyes, face or presence emotionally or figuratively (vacant,
hollow, dazed, burning, lifeless gaze and the like), it describes an
ACTOR'S PERFORMANCE captured by a real camera — realize it ONLY
through gaze direction, focus, eyelids, facial muscles, stillness
and posture. Every living person's eyes remain anatomically normal
human eyes with a natural iris and pupil, natural sclera and normal
proportions; NEVER whiten, blank out, cloud over, glow, enlarge or
otherwise alter eyeballs, skin or anatomy — unless the story
explicitly declares that being non-human or supernatural in form.

PROPS FACE THE RIGHT WAY: every handheld or used object must be
oriented exactly as its real-world use requires. A person reading,
watching or operating something (a phone, a photograph, a paper,
any device) has its functional side — screen, front, page — facing
THEIR OWN eyes; the camera then sees whatever side the staging
geometry implies (often its back). Show the functional side to the
camera ONLY when the shot text itself stages it toward the viewer.
Never flip, mirror or reverse an object's front and back.

CARRIED STATE (persist exactly — must match the neighbouring shots of this scene): Su-ri-young retains the folded coordinate note in her pocket and keeps her doll-adorned backpack beside her while checking the location.

PEOPLE: the SHOT TEXT alone decides whether any person is visible in this shot. IF a person appears, they must be one of: 수리영 (한국인 20대 초반 여성, 긴 검은 생머리, 젊은 타원형 얼굴) — never anyone else, and never add a person the shot text does not show. IF only part of a person is in frame (a hand, arm, foot, back, silhouette), that body part belongs to the specific person the shot text names — its sex, age, build, skin and grooming must unmistakably match that person's profile above.

No text, captions, watermarks or annotations anywhere.

Create ONE FINAL photorealistic live-action film still of the moment below — contemporary South Korea, 2026; all people are Korean unless stated. TIME OF DAY (lock): late morning.

SHOT TEXT (authoritative, Korean): 어두운 PC방 안, 수리영이 켜진 모니터 화면을 응시하며 앉아 있는 상체

LOCATION (lock): At a computer booth inside a dark internet café, illuminated mainly by the active monitor. The shot takes place here — the attached LOCATION PHOTOGRAPH shows the exact spot.

CAMERA & FRAME (follow exactly — this is the composition authority for this still; the reference images supply identity and place, never the framing):
- CAMERA: Begin at seated eye level in front of and slightly to one side of Suriyoung, showing her upper body on one half of the medium frame and the active monitor on the other. Start the prescribed slow dolly toward the desk without yet lowering, using her three-quarter posture and fixed eyeline to bind her to the coordinate search.
- FRAMING SCALE: medium shot
- KEY BACKGROUND ELEMENTS: PC방 모니터 (On and displaying the location search) — The active front face angles toward Suriyoung and remains partly readable to the camera as a coordinate-location display; used as Shares the frame with Suriyoung as the immediate focus of her search; PC방 책상 (In use) — Its front edge angles across the lower frame toward the monitor; used as Creates the working plane that the following downward camera move will approach.
Compose the frame exactly as specified above — camera angle, subject scale and screen placement. Keep true physical scale between people and background elements: fixtures and distant objects occupy only the small screen area their real size and distance dictate; never enlarge a background object into a foreground presence.

REALIZE FIGURATIVE LANGUAGE AS A LIVE-ACTION SHOT: the Korean shot
text may describe characters metaphorically, figuratively or with
exaggeration. Photograph what a real movie camera would actually
record on a physical set — exaggerated or figurative impressions
become realistic staging choices (lighting, distance, angle,
wardrobe), not literal fantasy imagery.
EVERY CHARACTER IS A HUMAN BEING: unless the story explicitly
features non-human or virtual beings (as in science fiction or
fantasy), every character — however indirectly, vaguely or
figuratively the text describes them — IS a real human. When the
text gives no direct visual description of a person, IMAGINE one
and still show them as a concrete, fully-formed human being:
always render the human form to the maximum extent — build,
posture, face, hands, clothing — never reduce a person to a shape,
blob, solid silhouette or abstract mass.

EXPRESSIONS ARE ACTED, NEVER ANATOMICAL: when the text describes a
person's eyes, face or presence emotionally or figuratively (vacant,
hollow, dazed, burning, lifeless gaze and the like), it describes an
ACTOR'S PERFORMANCE captured by a real camera — realize it ONLY
through gaze direction, focus, eyelids, facial muscles, stillness
and posture. Every living person's eyes remain anatomically normal
human eyes with a natural iris and pupil, natural sclera and normal
proportions; NEVER whiten, blank out, cloud over, glow, enlarge or
otherwise alter eyeballs, skin or anatomy — unless the story
explicitly declares that being non-human or supernatural in form.

PROPS FACE THE RIGHT WAY: every handheld or used object must be
oriented exactly as its real-world use requires. A person reading,
watching or operating something (a phone, a photograph, a paper,
any device) has its functional side — screen, front, page — facing
THEIR OWN eyes; the camera then sees whatever side the staging
geometry implies (often its back). Show the functional side to the
camera ONLY when the shot text itself stages it toward the viewer.
Never flip, mirror or reverse an object's front and back.

CARRIED STATE (persist exactly — must match the neighbouring shots of this scene): Su-ri-young retains the folded coordinate note in her pocket and keeps her doll-adorned backpack beside her while checking the location.

PEOPLE: the SHOT TEXT alone decides whether any person is visible in this shot. IF a person appears, they must be one of: 수리영 (한국인 20대 초반 여성, 긴 검은 생머리, 젊은 타원형 얼굴) — never anyone else, and never add a person the shot text does not show. IF only part of a person is in frame (a hand, arm, foot, back, silhouette), that body part belongs to the specific person the shot text names — its sex, age, build, skin and grooming must unmistakably match that person's profile above.

No text, captions, watermarks or annotations anywhere.

VISUAL APPROACH — this take (camera, framing, composition):
Use an intimate, gently compressed lens feel with selective focus centered on Suriyoung’s eyes and restrained facial tension, while retaining enough screen clarity for the coordinate-location display to remain partly readable. Let the monitor cast a cool, directional key across her face and upper body, with the surrounding café falling into soft darkness; the slight forward dolly creates quiet psychological pressure without changing camera height.

PRIORITY OF INSTRUCTIONS: the VISUAL APPROACH above governs camera,
framing and composition ONLY. Every other instruction in this prompt —
location, time of day, people and their identity, poses, props,
carried state, physical rules, and the no-text rule — is binding
exactly as written earlier. If the VISUAL APPROACH appears to conflict
with any of them, the earlier instruction wins.

Create ONE FINAL photorealistic live-action film still of the moment below — contemporary South Korea, 2026; all people are Korean unless stated. TIME OF DAY (lock): late morning.

SHOT TEXT (authoritative, Korean): 어두운 PC방 안, 수리영이 켜진 모니터 화면을 응시하며 앉아 있는 상체

LOCATION (lock): At a computer booth inside a dark internet café, illuminated mainly by the active monitor. The shot takes place here — the attached LOCATION PHOTOGRAPH shows the exact spot.

CAMERA & FRAME (follow exactly — this is the composition authority for this still; the reference images supply identity and place, never the framing):
- CAMERA: Begin at seated eye level in front of and slightly to one side of Suriyoung, showing her upper body on one half of the medium frame and the active monitor on the other. Start the prescribed slow dolly toward the desk without yet lowering, using her three-quarter posture and fixed eyeline to bind her to the coordinate search.
- FRAMING SCALE: medium shot
- KEY BACKGROUND ELEMENTS: PC방 모니터 (On and displaying the location search) — The active front face angles toward Suriyoung and remains partly readable to the camera as a coordinate-location display; used as Shares the frame with Suriyoung as the immediate focus of her search; PC방 책상 (In use) — Its front edge angles across the lower frame toward the monitor; used as Creates the working plane that the following downward camera move will approach.
Compose the frame exactly as specified above — camera angle, subject scale and screen placement. Keep true physical scale between people and background elements: fixtures and distant objects occupy only the small screen area their real size and distance dictate; never enlarge a background object into a foreground presence.

REALIZE FIGURATIVE LANGUAGE AS A LIVE-ACTION SHOT: the Korean shot
text may describe characters metaphorically, figuratively or with
exaggeration. Photograph what a real movie camera would actually
record on a physical set — exaggerated or figurative impressions
become realistic staging choices (lighting, distance, angle,
wardrobe), not literal fantasy imagery.
EVERY CHARACTER IS A HUMAN BEING: unless the story explicitly
features non-human or virtual beings (as in science fiction or
fantasy), every character — however indirectly, vaguely or
figuratively the text describes them — IS a real human. When the
text gives no direct visual description of a person, IMAGINE one
and still show them as a concrete, fully-formed human being:
always render the human form to the maximum extent — build,
posture, face, hands, clothing — never reduce a person to a shape,
blob, solid silhouette or abstract mass.

EXPRESSIONS ARE ACTED, NEVER ANATOMICAL: when the text describes a
person's eyes, face or presence emotionally or figuratively (vacant,
hollow, dazed, burning, lifeless gaze and the like), it describes an
ACTOR'S PERFORMANCE captured by a real camera — realize it ONLY
through gaze direction, focus, eyelids, facial muscles, stillness
and posture. Every living person's eyes remain anatomically normal
human eyes with a natural iris and pupil, natural sclera and normal
proportions; NEVER whiten, blank out, cloud over, glow, enlarge or
otherwise alter eyeballs, skin or anatomy — unless the story
explicitly declares that being non-human or supernatural in form.

PROPS FACE THE RIGHT WAY: every handheld or used object must be
oriented exactly as its real-world use requires. A person reading,
watching or operating something (a phone, a photograph, a paper,
any device) has its functional side — screen, front, page — facing
THEIR OWN eyes; the camera then sees whatever side the staging
geometry implies (often its back). Show the functional side to the
camera ONLY when the shot text itself stages it toward the viewer.
Never flip, mirror or reverse an object's front and back.

CARRIED STATE (persist exactly — must match the neighbouring shots of this scene): Su-ri-young retains the folded coordinate note in her pocket and keeps her doll-adorned backpack beside her while checking the location.

PEOPLE: the SHOT TEXT alone decides whether any person is visible in this shot. IF a person appears, they must be one of: 수리영 (한국인 20대 초반 여성, 긴 검은 생머리, 젊은 타원형 얼굴) — never anyone else, and never add a person the shot text does not show. IF only part of a person is in frame (a hand, arm, foot, back, silhouette), that body part belongs to the specific person the shot text names — its sex, age, build, skin and grooming must unmistakably match that person's profile above.

No text, captions, watermarks or annotations anywhere.

VISUAL APPROACH — this take (camera, framing, composition):
Use a more spatial, moderately wide lens feel and generous depth of field, holding Suriyoung and the angled monitor as equally legible halves of a firm two-plane composition. Keep the desk edge crisp as a diagonal leading toward the screen, while subdued late-morning ambient traces remain secondary to the monitor illumination; the level forward dolly should feel investigative, emphasizing the physical geometry between her fixed gaze and the coordinate display.

PRIORITY OF INSTRUCTIONS: the VISUAL APPROACH above governs camera,
framing and composition ONLY. Every other instruction in this prompt —
location, time of day, people and their identity, poses, props,
carried state, physical rules, and the no-text rule — is binding
exactly as written earlier. If the VISUAL APPROACH appears to conflict
with any of them, the earlier instruction wins.

Edit this photograph. Apply ONLY the corrections listed below,
each in place, changing nothing else:

CORRECTIONS:
- Remove all text from the monitor screen, leaving only the map and graphical UI elements.

PRESERVE EVERYTHING ELSE EXACTLY as in the original photograph —
every person, pose, face, garment, object, set feature, lighting,
camera and framing stay identical. No text, captions or
watermarks anywhere.

Create ONE FINAL photorealistic live-action film still of the moment below — contemporary South Korea, 2026; all people are Korean unless stated. TIME OF DAY (lock): late afternoon, stormy.

SHOT TEXT (authoritative, Korean): 빗물에 젖은 수리영이 입술을 꽉 깨문 채 앞을 노려보는 얼굴 클로즈업

LOCATION (lock): On an exposed coastal road beneath storm clouds, in heavy rain and lightning. The shot takes place here — the attached LOCATION PHOTOGRAPH shows the exact spot.

CAMERA & FRAME (follow exactly — this is the composition authority for this still; the reference images supply identity and place, never the framing):
- CAMERA: Track at Suriyoung's face height directly ahead and slightly off the bicycle's forward axis, matching her speed as the move resolves into a tight close-up. Her rain-soaked face occupies most of the frame in a slight three-quarter view, bitten lips centered and forward gaze held beyond camera, while only a restrained trace of motion remains at the frame edges.
- FRAMING SCALE: close-up
- KEY BACKGROUND ELEMENTS: 세찬 비 (Falling heavily); used as Moving background texture that confirms the severe weather around her face; 해안가 (Visible only as blurred surroundings); used as Distant environmental context kept soft behind the moving close-up; 먹구름과 번개 (Storm clouds present with lightning); used as Upper-background weather context with intermittent visual punctuation.
Compose the frame exactly as specified above — camera angle, subject scale and screen placement. Keep true physical scale between people and background elements: fixtures and distant objects occupy only the small screen area their real size and distance dictate; never enlarge a background object into a foreground presence.

REALIZE FIGURATIVE LANGUAGE AS A LIVE-ACTION SHOT: the Korean shot
text may describe characters metaphorically, figuratively or with
exaggeration. Photograph what a real movie camera would actually
record on a physical set — exaggerated or figurative impressions
become realistic staging choices (lighting, distance, angle,
wardrobe), not literal fantasy imagery.
EVERY CHARACTER IS A HUMAN BEING: unless the story explicitly
features non-human or virtual beings (as in science fiction or
fantasy), every character — however indirectly, vaguely or
figuratively the text describes them — IS a real human. When the
text gives no direct visual description of a person, IMAGINE one
and still show them as a concrete, fully-formed human being:
always render the human form to the maximum extent — build,
posture, face, hands, clothing — never reduce a person to a shape,
blob, solid silhouette or abstract mass.

EXPRESSIONS ARE ACTED, NEVER ANATOMICAL: when the text describes a
person's eyes, face or presence emotionally or figuratively (vacant,
hollow, dazed, burning, lifeless gaze and the like), it describes an
ACTOR'S PERFORMANCE captured by a real camera — realize it ONLY
through gaze direction, focus, eyelids, facial muscles, stillness
and posture. Every living person's eyes remain anatomically normal
human eyes with a natural iris and pupil, natural sclera and normal
proportions; NEVER whiten, blank out, cloud over, glow, enlarge or
otherwise alter eyeballs, skin or anatomy — unless the story
explicitly declares that being non-human or supernatural in form.

PROPS FACE THE RIGHT WAY: every handheld or used object must be
oriented exactly as its real-world use requires. A person reading,
watching or operating something (a phone, a photograph, a paper,
any device) has its functional side — screen, front, page — facing
THEIR OWN eyes; the camera then sees whatever side the staging
geometry implies (often its back). Show the functional side to the
camera ONLY when the shot text itself stages it toward the viewer.
Never flip, mirror or reverse an object's front and back.

CARRIED STATE (persist exactly — must match the neighbouring shots of this scene): Su-ri-young rides the same bicycle through the rain with her backpack containing the folded maps; the coordinate note and phone remain with her.

PEOPLE: the SHOT TEXT alone decides whether any person is visible in this shot. IF a person appears, they must be one of: 수리영 (한국인 20대 초반 여성, 긴 검은 생머리, 젊은 타원형 얼굴) — never anyone else, and never add a person the shot text does not show. IF only part of a person is in frame (a hand, arm, foot, back, silhouette), that body part belongs to the specific person the shot text names — its sex, age, build, skin and grooming must unmistakably match that person's profile above.

No text, captions, watermarks or annotations anywhere.

Create ONE FINAL photorealistic live-action film still of the moment below — contemporary South Korea, 2026; all people are Korean unless stated. TIME OF DAY (lock): late afternoon, stormy.

SHOT TEXT (authoritative, Korean): 빗물에 젖은 수리영이 입술을 꽉 깨문 채 앞을 노려보는 얼굴 클로즈업

LOCATION (lock): On an exposed coastal road beneath storm clouds, in heavy rain and lightning. The shot takes place here — the attached LOCATION PHOTOGRAPH shows the exact spot.

CAMERA & FRAME (follow exactly — this is the composition authority for this still; the reference images supply identity and place, never the framing):
- CAMERA: Track at Suriyoung's face height directly ahead and slightly off the bicycle's forward axis, matching her speed as the move resolves into a tight close-up. Her rain-soaked face occupies most of the frame in a slight three-quarter view, bitten lips centered and forward gaze held beyond camera, while only a restrained trace of motion remains at the frame edges.
- FRAMING SCALE: close-up
- KEY BACKGROUND ELEMENTS: 세찬 비 (Falling heavily); used as Moving background texture that confirms the severe weather around her face; 해안가 (Visible only as blurred surroundings); used as Distant environmental context kept soft behind the moving close-up; 먹구름과 번개 (Storm clouds present with lightning); used as Upper-background weather context with intermittent visual punctuation.
Compose the frame exactly as specified above — camera angle, subject scale and screen placement. Keep true physical scale between people and background elements: fixtures and distant objects occupy only the small screen area their real size and distance dictate; never enlarge a background object into a foreground presence.

REALIZE FIGURATIVE LANGUAGE AS A LIVE-ACTION SHOT: the Korean shot
text may describe characters metaphorically, figuratively or with
exaggeration. Photograph what a real movie camera would actually
record on a physical set — exaggerated or figurative impressions
become realistic staging choices (lighting, distance, angle,
wardrobe), not literal fantasy imagery.
EVERY CHARACTER IS A HUMAN BEING: unless the story explicitly
features non-human or virtual beings (as in science fiction or
fantasy), every character — however indirectly, vaguely or
figuratively the text describes them — IS a real human. When the
text gives no direct visual description of a person, IMAGINE one
and still show them as a concrete, fully-formed human being:
always render the human form to the maximum extent — build,
posture, face, hands, clothing — never reduce a person to a shape,
blob, solid silhouette or abstract mass.

EXPRESSIONS ARE ACTED, NEVER ANATOMICAL: when the text describes a
person's eyes, face or presence emotionally or figuratively (vacant,
hollow, dazed, burning, lifeless gaze and the like), it describes an
ACTOR'S PERFORMANCE captured by a real camera — realize it ONLY
through gaze direction, focus, eyelids, facial muscles, stillness
and posture. Every living person's eyes remain anatomically normal
human eyes with a natural iris and pupil, natural sclera and normal
proportions; NEVER whiten, blank out, cloud over, glow, enlarge or
otherwise alter eyeballs, skin or anatomy — unless the story
explicitly declares that being non-human or supernatural in form.

PROPS FACE THE RIGHT WAY: every handheld or used object must be
oriented exactly as its real-world use requires. A person reading,
watching or operating something (a phone, a photograph, a paper,
any device) has its functional side — screen, front, page — facing
THEIR OWN eyes; the camera then sees whatever side the staging
geometry implies (often its back). Show the functional side to the
camera ONLY when the shot text itself stages it toward the viewer.
Never flip, mirror or reverse an object's front and back.

CARRIED STATE (persist exactly — must match the neighbouring shots of this scene): Su-ri-young rides the same bicycle through the rain with her backpack containing the folded maps; the coordinate note and phone remain with her.

PEOPLE: the SHOT TEXT alone decides whether any person is visible in this shot. IF a person appears, they must be one of: 수리영 (한국인 20대 초반 여성, 긴 검은 생머리, 젊은 타원형 얼굴) — never anyone else, and never add a person the shot text does not show. IF only part of a person is in frame (a hand, arm, foot, back, silhouette), that body part belongs to the specific person the shot text names — its sex, age, build, skin and grooming must unmistakably match that person's profile above.

No text, captions, watermarks or annotations anywhere.

VISUAL APPROACH — this take (camera, framing, composition):
Use an intimate, compressed lens feel with extremely shallow focus, holding the eyes and tightly bitten lips in crisp detail while the rain, coast, and storm layers dissolve into soft movement behind her. Keep the tracking nearly frictionless at face height, with only subtle edge smear conveying speed. Shape the wet face with cool, diffuse late-afternoon storm light, allowing a lightning pulse to create a brief fine highlight along the rain and hair without overpowering her expression.

PRIORITY OF INSTRUCTIONS: the VISUAL APPROACH above governs camera,
framing and composition ONLY. Every other instruction in this prompt —
location, time of day, people and their identity, poses, props,
carried state, physical rules, and the no-text rule — is binding
exactly as written earlier. If the VISUAL APPROACH appears to conflict
with any of them, the earlier instruction wins.

Create ONE FINAL photorealistic live-action film still of the moment below — contemporary South Korea, 2026; all people are Korean unless stated. TIME OF DAY (lock): late afternoon, stormy.

SHOT TEXT (authoritative, Korean): 빗물에 젖은 수리영이 입술을 꽉 깨문 채 앞을 노려보는 얼굴 클로즈업

LOCATION (lock): On an exposed coastal road beneath storm clouds, in heavy rain and lightning. The shot takes place here — the attached LOCATION PHOTOGRAPH shows the exact spot.

CAMERA & FRAME (follow exactly — this is the composition authority for this still; the reference images supply identity and place, never the framing):
- CAMERA: Track at Suriyoung's face height directly ahead and slightly off the bicycle's forward axis, matching her speed as the move resolves into a tight close-up. Her rain-soaked face occupies most of the frame in a slight three-quarter view, bitten lips centered and forward gaze held beyond camera, while only a restrained trace of motion remains at the frame edges.
- FRAMING SCALE: close-up
- KEY BACKGROUND ELEMENTS: 세찬 비 (Falling heavily); used as Moving background texture that confirms the severe weather around her face; 해안가 (Visible only as blurred surroundings); used as Distant environmental context kept soft behind the moving close-up; 먹구름과 번개 (Storm clouds present with lightning); used as Upper-background weather context with intermittent visual punctuation.
Compose the frame exactly as specified above — camera angle, subject scale and screen placement. Keep true physical scale between people and background elements: fixtures and distant objects occupy only the small screen area their real size and distance dictate; never enlarge a background object into a foreground presence.

REALIZE FIGURATIVE LANGUAGE AS A LIVE-ACTION SHOT: the Korean shot
text may describe characters metaphorically, figuratively or with
exaggeration. Photograph what a real movie camera would actually
record on a physical set — exaggerated or figurative impressions
become realistic staging choices (lighting, distance, angle,
wardrobe), not literal fantasy imagery.
EVERY CHARACTER IS A HUMAN BEING: unless the story explicitly
features non-human or virtual beings (as in science fiction or
fantasy), every character — however indirectly, vaguely or
figuratively the text describes them — IS a real human. When the
text gives no direct visual description of a person, IMAGINE one
and still show them as a concrete, fully-formed human being:
always render the human form to the maximum extent — build,
posture, face, hands, clothing — never reduce a person to a shape,
blob, solid silhouette or abstract mass.

EXPRESSIONS ARE ACTED, NEVER ANATOMICAL: when the text describes a
person's eyes, face or presence emotionally or figuratively (vacant,
hollow, dazed, burning, lifeless gaze and the like), it describes an
ACTOR'S PERFORMANCE captured by a real camera — realize it ONLY
through gaze direction, focus, eyelids, facial muscles, stillness
and posture. Every living person's eyes remain anatomically normal
human eyes with a natural iris and pupil, natural sclera and normal
proportions; NEVER whiten, blank out, cloud over, glow, enlarge or
otherwise alter eyeballs, skin or anatomy — unless the story
explicitly declares that being non-human or supernatural in form.

PROPS FACE THE RIGHT WAY: every handheld or used object must be
oriented exactly as its real-world use requires. A person reading,
watching or operating something (a phone, a photograph, a paper,
any device) has its functional side — screen, front, page — facing
THEIR OWN eyes; the camera then sees whatever side the staging
geometry implies (often its back). Show the functional side to the
camera ONLY when the shot text itself stages it toward the viewer.
Never flip, mirror or reverse an object's front and back.

CARRIED STATE (persist exactly — must match the neighbouring shots of this scene): Su-ri-young rides the same bicycle through the rain with her backpack containing the folded maps; the coordinate note and phone remain with her.

PEOPLE: the SHOT TEXT alone decides whether any person is visible in this shot. IF a person appears, they must be one of: 수리영 (한국인 20대 초반 여성, 긴 검은 생머리, 젊은 타원형 얼굴) — never anyone else, and never add a person the shot text does not show. IF only part of a person is in frame (a hand, arm, foot, back, silhouette), that body part belongs to the specific person the shot text names — its sex, age, build, skin and grooming must unmistakably match that person's profile above.

No text, captions, watermarks or annotations anywhere.

VISUAL APPROACH — this take (camera, framing, composition):
Use a close, moderately wide lens feel that gives the face strong dimensional presence while stretching the blurred rain and coastal depth behind it. Organize the heavy rain into diagonal foreground and background streaks, with restrained peripheral motion making the forward drive feel urgent while the centered mouth and fixed gaze remain visually stable. Preserve texture across the soaked skin and hair under broad storm light, using the upper-background lightning as a sharp, intermittent punctuation within the dark cloud mass.

PRIORITY OF INSTRUCTIONS: the VISUAL APPROACH above governs camera,
framing and composition ONLY. Every other instruction in this prompt —
location, time of day, people and their identity, poses, props,
carried state, physical rules, and the no-text rule — is binding
exactly as written earlier. If the VISUAL APPROACH appears to conflict
with any of them, the earlier instruction wins.

Edit this photograph. Apply ONLY the corrections listed below,
each in place, changing nothing else:

CORRECTIONS:
- Turn the character's head to a slight three-quarter view rather than a direct frontal angle.

PRESERVE EVERYTHING ELSE EXACTLY as in the original photograph —
every person, pose, face, garment, object, set feature, lighting,
camera and framing stay identical. No text, captions or
watermarks anywhere.

Create ONE FINAL photorealistic live-action film still of the moment below — contemporary South Korea, 2026; all people are Korean unless stated. TIME OF DAY (lock): dusk, foggy.

SHOT TEXT (authoritative, Korean): 수리영이 축축한 바닷가 갯바위에 무릎을 모아 안고 웅크려 앉아 있는 전신

LOCATION (lock): On damp rocks at the edge of an empty harbor, facing the fog-covered sea at dusk. The shot takes place here — the attached LOCATION PHOTOGRAPH shows the exact spot.

CAMERA & FRAME (follow exactly — this is the composition authority for this still; the reference images supply identity and place, never the framing):
- CAMERA: From seated eye level in front of 수리영, the static wide frame places her compact full figure in the lower middle against the empty harbor and the water she watches. Her body is turned slightly away from a direct frontal portrait, while the rocks provide a grounded foreground and fog reduces the distant spatial layers.
- FRAMING SCALE: wide shot
- FRAME LAYOUT: 수리영 in the lower-center of the frame, midground, looks toward open water; open water in the upper-center of the frame, background.
- KEY BACKGROUND ELEMENTS: Harbor rocks (Wet after the shower); used as Foreground base beneath 수리영's seated figure; Harbor water (Partly obscured in the distance); used as Open background toward which 수리영 directs her attention; Empty harbor (Empty); used as Negative space emphasizing her isolation and exhaustion; Fog (Dense over the harbor); used as Background depth suppression that conceals distant boats.
Compose the frame exactly as specified above — camera angle, subject scale and screen placement. Keep true physical scale between people and background elements: fixtures and distant objects occupy only the small screen area their real size and distance dictate; never enlarge a background object into a foreground presence.

REALIZE FIGURATIVE LANGUAGE AS A LIVE-ACTION SHOT: the Korean shot
text may describe characters metaphorically, figuratively or with
exaggeration. Photograph what a real movie camera would actually
record on a physical set — exaggerated or figurative impressions
become realistic staging choices (lighting, distance, angle,
wardrobe), not literal fantasy imagery.
EVERY CHARACTER IS A HUMAN BEING: unless the story explicitly
features non-human or virtual beings (as in science fiction or
fantasy), every character — however indirectly, vaguely or
figuratively the text describes them — IS a real human. When the
text gives no direct visual description of a person, IMAGINE one
and still show them as a concrete, fully-formed human being:
always render the human form to the maximum extent — build,
posture, face, hands, clothing — never reduce a person to a shape,
blob, solid silhouette or abstract mass.

EXPRESSIONS ARE ACTED, NEVER ANATOMICAL: when the text describes a
person's eyes, face or presence emotionally or figuratively (vacant,
hollow, dazed, burning, lifeless gaze and the like), it describes an
ACTOR'S PERFORMANCE captured by a real camera — realize it ONLY
through gaze direction, focus, eyelids, facial muscles, stillness
and posture. Every living person's eyes remain anatomically normal
human eyes with a natural iris and pupil, natural sclera and normal
proportions; NEVER whiten, blank out, cloud over, glow, enlarge or
otherwise alter eyeballs, skin or anatomy — unless the story
explicitly declares that being non-human or supernatural in form.

PROPS FACE THE RIGHT WAY: every handheld or used object must be
oriented exactly as its real-world use requires. A person reading,
watching or operating something (a phone, a photograph, a paper,
any device) has its functional side — screen, front, page — facing
THEIR OWN eyes; the camera then sees whatever side the staging
geometry implies (often its back). Show the functional side to the
camera ONLY when the shot text itself stages it toward the viewer.
Never flip, mirror or reverse an object's front and back.

CARRIED STATE (persist exactly — must match the neighbouring shots of this scene): Su-ri-young still has the old photograph, her map-filled backpack, the coordinate note, and her phone with its repeated missed calls from Hye-su.

PEOPLE: the SHOT TEXT alone decides whether any person is visible in this shot. IF a person appears, they must be one of: 수리영 (한국인 20대 초반 여성, 긴 검은 생머리, 젊은 타원형 얼굴) — never anyone else, and never add a person the shot text does not show. IF only part of a person is in frame (a hand, arm, foot, back, silhouette), that body part belongs to the specific person the shot text names — its sex, age, build, skin and grooming must unmistakably match that person's profile above.

No text, captions, watermarks or annotations anywhere.

Create ONE FINAL photorealistic live-action film still of the moment below — contemporary South Korea, 2026; all people are Korean unless stated. TIME OF DAY (lock): dusk, foggy.

SHOT TEXT (authoritative, Korean): 수리영이 축축한 바닷가 갯바위에 무릎을 모아 안고 웅크려 앉아 있는 전신

LOCATION (lock): On damp rocks at the edge of an empty harbor, facing the fog-covered sea at dusk. The shot takes place here — the attached LOCATION PHOTOGRAPH shows the exact spot.

CAMERA & FRAME (follow exactly — this is the composition authority for this still; the reference images supply identity and place, never the framing):
- CAMERA: From seated eye level in front of 수리영, the static wide frame places her compact full figure in the lower middle against the empty harbor and the water she watches. Her body is turned slightly away from a direct frontal portrait, while the rocks provide a grounded foreground and fog reduces the distant spatial layers.
- FRAMING SCALE: wide shot
- FRAME LAYOUT: 수리영 in the lower-center of the frame, midground, looks toward open water; open water in the upper-center of the frame, background.
- KEY BACKGROUND ELEMENTS: Harbor rocks (Wet after the shower); used as Foreground base beneath 수리영's seated figure; Harbor water (Partly obscured in the distance); used as Open background toward which 수리영 directs her attention; Empty harbor (Empty); used as Negative space emphasizing her isolation and exhaustion; Fog (Dense over the harbor); used as Background depth suppression that conceals distant boats.
Compose the frame exactly as specified above — camera angle, subject scale and screen placement. Keep true physical scale between people and background elements: fixtures and distant objects occupy only the small screen area their real size and distance dictate; never enlarge a background object into a foreground presence.

REALIZE FIGURATIVE LANGUAGE AS A LIVE-ACTION SHOT: the Korean shot
text may describe characters metaphorically, figuratively or with
exaggeration. Photograph what a real movie camera would actually
record on a physical set — exaggerated or figurative impressions
become realistic staging choices (lighting, distance, angle,
wardrobe), not literal fantasy imagery.
EVERY CHARACTER IS A HUMAN BEING: unless the story explicitly
features non-human or virtual beings (as in science fiction or
fantasy), every character — however indirectly, vaguely or
figuratively the text describes them — IS a real human. When the
text gives no direct visual description of a person, IMAGINE one
and still show them as a concrete, fully-formed human being:
always render the human form to the maximum extent — build,
posture, face, hands, clothing — never reduce a person to a shape,
blob, solid silhouette or abstract mass.

EXPRESSIONS ARE ACTED, NEVER ANATOMICAL: when the text describes a
person's eyes, face or presence emotionally or figuratively (vacant,
hollow, dazed, burning, lifeless gaze and the like), it describes an
ACTOR'S PERFORMANCE captured by a real camera — realize it ONLY
through gaze direction, focus, eyelids, facial muscles, stillness
and posture. Every living person's eyes remain anatomically normal
human eyes with a natural iris and pupil, natural sclera and normal
proportions; NEVER whiten, blank out, cloud over, glow, enlarge or
otherwise alter eyeballs, skin or anatomy — unless the story
explicitly declares that being non-human or supernatural in form.

PROPS FACE THE RIGHT WAY: every handheld or used object must be
oriented exactly as its real-world use requires. A person reading,
watching or operating something (a phone, a photograph, a paper,
any device) has its functional side — screen, front, page — facing
THEIR OWN eyes; the camera then sees whatever side the staging
geometry implies (often its back). Show the functional side to the
camera ONLY when the shot text itself stages it toward the viewer.
Never flip, mirror or reverse an object's front and back.

CARRIED STATE (persist exactly — must match the neighbouring shots of this scene): Su-ri-young still has the old photograph, her map-filled backpack, the coordinate note, and her phone with its repeated missed calls from Hye-su.

PEOPLE: the SHOT TEXT alone decides whether any person is visible in this shot. IF a person appears, they must be one of: 수리영 (한국인 20대 초반 여성, 긴 검은 생머리, 젊은 타원형 얼굴) — never anyone else, and never add a person the shot text does not show. IF only part of a person is in frame (a hand, arm, foot, back, silhouette), that body part belongs to the specific person the shot text names — its sex, age, build, skin and grooming must unmistakably match that person's profile above.

No text, captions, watermarks or annotations anywhere.

VISUAL APPROACH — this take (camera, framing, composition):
Use a restrained wide-angle perspective with the camera held at the prescribed seated eye level, keeping the full figure compact and the foreground rocks, midground subject, and fog-softened water clearly layered. Favor deep focus and broad negative space, with soft, cool dusk illumination diffused evenly through the fog and subtle wet reflections on the rocks. The composition should feel still, exposed, and quietly observational.

PRIORITY OF INSTRUCTIONS: the VISUAL APPROACH above governs camera,
framing and composition ONLY. Every other instruction in this prompt —
location, time of day, people and their identity, poses, props,
carried state, physical rules, and the no-text rule — is binding
exactly as written earlier. If the VISUAL APPROACH appears to conflict
with any of them, the earlier instruction wins.

Create ONE FINAL photorealistic live-action film still of the moment below — contemporary South Korea, 2026; all people are Korean unless stated. TIME OF DAY (lock): dusk, foggy.

SHOT TEXT (authoritative, Korean): 수리영이 축축한 바닷가 갯바위에 무릎을 모아 안고 웅크려 앉아 있는 전신

LOCATION (lock): On damp rocks at the edge of an empty harbor, facing the fog-covered sea at dusk. The shot takes place here — the attached LOCATION PHOTOGRAPH shows the exact spot.

CAMERA & FRAME (follow exactly — this is the composition authority for this still; the reference images supply identity and place, never the framing):
- CAMERA: From seated eye level in front of 수리영, the static wide frame places her compact full figure in the lower middle against the empty harbor and the water she watches. Her body is turned slightly away from a direct frontal portrait, while the rocks provide a grounded foreground and fog reduces the distant spatial layers.
- FRAMING SCALE: wide shot
- FRAME LAYOUT: 수리영 in the lower-center of the frame, midground, looks toward open water; open water in the upper-center of the frame, background.
- KEY BACKGROUND ELEMENTS: Harbor rocks (Wet after the shower); used as Foreground base beneath 수리영's seated figure; Harbor water (Partly obscured in the distance); used as Open background toward which 수리영 directs her attention; Empty harbor (Empty); used as Negative space emphasizing her isolation and exhaustion; Fog (Dense over the harbor); used as Background depth suppression that conceals distant boats.
Compose the frame exactly as specified above — camera angle, subject scale and screen placement. Keep true physical scale between people and background elements: fixtures and distant objects occupy only the small screen area their real size and distance dictate; never enlarge a background object into a foreground presence.

REALIZE FIGURATIVE LANGUAGE AS A LIVE-ACTION SHOT: the Korean shot
text may describe characters metaphorically, figuratively or with
exaggeration. Photograph what a real movie camera would actually
record on a physical set — exaggerated or figurative impressions
become realistic staging choices (lighting, distance, angle,
wardrobe), not literal fantasy imagery.
EVERY CHARACTER IS A HUMAN BEING: unless the story explicitly
features non-human or virtual beings (as in science fiction or
fantasy), every character — however indirectly, vaguely or
figuratively the text describes them — IS a real human. When the
text gives no direct visual description of a person, IMAGINE one
and still show them as a concrete, fully-formed human being:
always render the human form to the maximum extent — build,
posture, face, hands, clothing — never reduce a person to a shape,
blob, solid silhouette or abstract mass.

EXPRESSIONS ARE ACTED, NEVER ANATOMICAL: when the text describes a
person's eyes, face or presence emotionally or figuratively (vacant,
hollow, dazed, burning, lifeless gaze and the like), it describes an
ACTOR'S PERFORMANCE captured by a real camera — realize it ONLY
through gaze direction, focus, eyelids, facial muscles, stillness
and posture. Every living person's eyes remain anatomically normal
human eyes with a natural iris and pupil, natural sclera and normal
proportions; NEVER whiten, blank out, cloud over, glow, enlarge or
otherwise alter eyeballs, skin or anatomy — unless the story
explicitly declares that being non-human or supernatural in form.

PROPS FACE THE RIGHT WAY: every handheld or used object must be
oriented exactly as its real-world use requires. A person reading,
watching or operating something (a phone, a photograph, a paper,
any device) has its functional side — screen, front, page — facing
THEIR OWN eyes; the camera then sees whatever side the staging
geometry implies (often its back). Show the functional side to the
camera ONLY when the shot text itself stages it toward the viewer.
Never flip, mirror or reverse an object's front and back.

CARRIED STATE (persist exactly — must match the neighbouring shots of this scene): Su-ri-young still has the old photograph, her map-filled backpack, the coordinate note, and her phone with its repeated missed calls from Hye-su.

PEOPLE: the SHOT TEXT alone decides whether any person is visible in this shot. IF a person appears, they must be one of: 수리영 (한국인 20대 초반 여성, 긴 검은 생머리, 젊은 타원형 얼굴) — never anyone else, and never add a person the shot text does not show. IF only part of a person is in frame (a hand, arm, foot, back, silhouette), that body part belongs to the specific person the shot text names — its sex, age, build, skin and grooming must unmistakably match that person's profile above.

No text, captions, watermarks or annotations anywhere.

VISUAL APPROACH — this take (camera, framing, composition):
Photograph from farther back with a compressed long-lens feel while preserving the prescribed wide framing, seated eye level, and lower-center placement. Flatten the rocks, full figure, and obscured water into sparse horizontal layers, letting the fog erase depth and make the surrounding emptiness feel dense and enclosing. Use subdued dusk contrast, with the damp rock surfaces slightly darker than the soft gray water, creating a hushed, emotionally compressed image.

PRIORITY OF INSTRUCTIONS: the VISUAL APPROACH above governs camera,
framing and composition ONLY. Every other instruction in this prompt —
location, time of day, people and their identity, poses, props,
carried state, physical rules, and the no-text rule — is binding
exactly as written earlier. If the VISUAL APPROACH appears to conflict
with any of them, the earlier instruction wins.

Edit this photograph. Apply ONLY the corrections listed below,
each in place, changing nothing else:

CORRECTIONS:
- Remove all distant boats from the background harbor to make it completely empty.

PRESERVE EVERYTHING ELSE EXACTLY as in the original photograph —
every person, pose, face, garment, object, set feature, lighting,
camera and framing stay identical. No text, captions or
watermarks anywhere.

Create ONE FINAL photorealistic live-action film still of the moment below — contemporary South Korea, 2026; all people are Korean unless stated. TIME OF DAY (lock): dusk.

SHOT TEXT (authoritative, Korean): 어두운 실루엣의 낯선 아시아인 남성이 숲속에서 습격당한 남자의 어깨를 강하게 덮치며 함께 바닥으로 쓰러지는 mid-impact 전신

LOCATION (lock): In a patch of dense woodland undergrowth at dusk, where the fleeing man is knocked to the forest floor. The shot takes place here — the attached PREVIOUS SHOT STILL shows this exact place.

CAMERA & FRAME (follow exactly — this is the composition authority for this still; the reference images supply identity and place, never the framing):
- CAMERA: At waist height and several meters away, the camera reaches the end of its rapid dolly-out in a wide, slightly canted lateral view, preserving both figures head to foot as their combined momentum drops toward the lower center. The fleeing man occupies middle-right while the dark incoming figure cuts across from middle-left, with the forest retained around their fall for scale.
- FRAMING SCALE: wide shot
- FRAME LAYOUT: 낯선 아시아인 남성 in the middle-left of the frame, midground, moves toward ground at lower center; 습격당한 남자 in the middle-right of the frame, midground, moves toward ground at lower center; impact ground in the lower-center of the frame, background.
- KEY BACKGROUND ELEMENTS: 우거진 수풀 (Dense forest growth); used as Encloses the collision and supplies full-body scale around the falling figures; 숲 바닥 (State not otherwise specified); used as Landing plane beneath the two falling bodies.
Compose the frame exactly as specified above — camera angle, subject scale and screen placement. Keep true physical scale between people and background elements: fixtures and distant objects occupy only the small screen area their real size and distance dictate; never enlarge a background object into a foreground presence.

THIS SHOT CONTINUES THE PREVIOUS SHOT: everything the attached still established — the place, its fixed features and wear, each person's clothing and state — persists. Any person in it who cannot move stays PRECISELY as photographed (body, pose, contact points, held objects); only the camera changes.

REALIZE FIGURATIVE LANGUAGE AS A LIVE-ACTION SHOT: the Korean shot
text may describe characters metaphorically, figuratively or with
exaggeration. Photograph what a real movie camera would actually
record on a physical set — exaggerated or figurative impressions
become realistic staging choices (lighting, distance, angle,
wardrobe), not literal fantasy imagery.
EVERY CHARACTER IS A HUMAN BEING: unless the story explicitly
features non-human or virtual beings (as in science fiction or
fantasy), every character — however indirectly, vaguely or
figuratively the text describes them — IS a real human. When the
text gives no direct visual description of a person, IMAGINE one
and still show them as a concrete, fully-formed human being:
always render the human form to the maximum extent — build,
posture, face, hands, clothing — never reduce a person to a shape,
blob, solid silhouette or abstract mass.

EXPRESSIONS ARE ACTED, NEVER ANATOMICAL: when the text describes a
person's eyes, face or presence emotionally or figuratively (vacant,
hollow, dazed, burning, lifeless gaze and the like), it describes an
ACTOR'S PERFORMANCE captured by a real camera — realize it ONLY
through gaze direction, focus, eyelids, facial muscles, stillness
and posture. Every living person's eyes remain anatomically normal
human eyes with a natural iris and pupil, natural sclera and normal
proportions; NEVER whiten, blank out, cloud over, glow, enlarge or
otherwise alter eyeballs, skin or anatomy — unless the story
explicitly declares that being non-human or supernatural in form.

PROPS FACE THE RIGHT WAY: every handheld or used object must be
oriented exactly as its real-world use requires. A person reading,
watching or operating something (a phone, a photograph, a paper,
any device) has its functional side — screen, front, page — facing
THEIR OWN eyes; the camera then sees whatever side the staging
geometry implies (often its back). Show the functional side to the
camera ONLY when the shot text itself stages it toward the viewer.
Never flip, mirror or reverse an object's front and back.

CARRIED STATE (persist exactly — must match the neighbouring shots of this scene): As he is knocked to the ground, the man still has one shoe missing and the bleeding bite mark on his thigh.

PEOPLE: the SHOT TEXT alone decides whether any person is visible in this shot. IF a person appears, they must be one of: 습격당한 남자 (한국인 성인 남성, 짧은 검은 머리, 자연스러운 얼굴형); 정체불명의 아시아인 남성 (포식 활성 상태) (동아시아계 성인 남성, 짧은 검은 머리, 자연스러운 남성 얼굴) — never anyone else, and never add a person the shot text does not show. IF only part of a person is in frame (a hand, arm, foot, back, silhouette), that body part belongs to the specific person the shot text names — its sex, age, build, skin and grooming must unmistakably match that person's profile above.

No text, captions, watermarks or annotations anywhere.

Create ONE FINAL photorealistic live-action film still of the moment below — contemporary South Korea, 2026; all people are Korean unless stated. TIME OF DAY (lock): dusk.

SHOT TEXT (authoritative, Korean): 어두운 실루엣의 낯선 아시아인 남성이 숲속에서 습격당한 남자의 어깨를 강하게 덮치며 함께 바닥으로 쓰러지는 mid-impact 전신

LOCATION (lock): In a patch of dense woodland undergrowth at dusk, where the fleeing man is knocked to the forest floor. The shot takes place here — the attached PREVIOUS SHOT STILL shows this exact place.

CAMERA & FRAME (follow exactly — this is the composition authority for this still; the reference images supply identity and place, never the framing):
- CAMERA: At waist height and several meters away, the camera reaches the end of its rapid dolly-out in a wide, slightly canted lateral view, preserving both figures head to foot as their combined momentum drops toward the lower center. The fleeing man occupies middle-right while the dark incoming figure cuts across from middle-left, with the forest retained around their fall for scale.
- FRAMING SCALE: wide shot
- FRAME LAYOUT: 낯선 아시아인 남성 in the middle-left of the frame, midground, moves toward ground at lower center; 습격당한 남자 in the middle-right of the frame, midground, moves toward ground at lower center; impact ground in the lower-center of the frame, background.
- KEY BACKGROUND ELEMENTS: 우거진 수풀 (Dense forest growth); used as Encloses the collision and supplies full-body scale around the falling figures; 숲 바닥 (State not otherwise specified); used as Landing plane beneath the two falling bodies.
Compose the frame exactly as specified above — camera angle, subject scale and screen placement. Keep true physical scale between people and background elements: fixtures and distant objects occupy only the small screen area their real size and distance dictate; never enlarge a background object into a foreground presence.

THIS SHOT CONTINUES THE PREVIOUS SHOT: everything the attached still established — the place, its fixed features and wear, each person's clothing and state — persists. Any person in it who cannot move stays PRECISELY as photographed (body, pose, contact points, held objects); only the camera changes.

REALIZE FIGURATIVE LANGUAGE AS A LIVE-ACTION SHOT: the Korean shot
text may describe characters metaphorically, figuratively or with
exaggeration. Photograph what a real movie camera would actually
record on a physical set — exaggerated or figurative impressions
become realistic staging choices (lighting, distance, angle,
wardrobe), not literal fantasy imagery.
EVERY CHARACTER IS A HUMAN BEING: unless the story explicitly
features non-human or virtual beings (as in science fiction or
fantasy), every character — however indirectly, vaguely or
figuratively the text describes them — IS a real human. When the
text gives no direct visual description of a person, IMAGINE one
and still show them as a concrete, fully-formed human being:
always render the human form to the maximum extent — build,
posture, face, hands, clothing — never reduce a person to a shape,
blob, solid silhouette or abstract mass.

EXPRESSIONS ARE ACTED, NEVER ANATOMICAL: when the text describes a
person's eyes, face or presence emotionally or figuratively (vacant,
hollow, dazed, burning, lifeless gaze and the like), it describes an
ACTOR'S PERFORMANCE captured by a real camera — realize it ONLY
through gaze direction, focus, eyelids, facial muscles, stillness
and posture. Every living person's eyes remain anatomically normal
human eyes with a natural iris and pupil, natural sclera and normal
proportions; NEVER whiten, blank out, cloud over, glow, enlarge or
otherwise alter eyeballs, skin or anatomy — unless the story
explicitly declares that being non-human or supernatural in form.

PROPS FACE THE RIGHT WAY: every handheld or used object must be
oriented exactly as its real-world use requires. A person reading,
watching or operating something (a phone, a photograph, a paper,
any device) has its functional side — screen, front, page — facing
THEIR OWN eyes; the camera then sees whatever side the staging
geometry implies (often its back). Show the functional side to the
camera ONLY when the shot text itself stages it toward the viewer.
Never flip, mirror or reverse an object's front and back.

CARRIED STATE (persist exactly — must match the neighbouring shots of this scene): As he is knocked to the ground, the man still has one shoe missing and the bleeding bite mark on his thigh.

PEOPLE: the SHOT TEXT alone decides whether any person is visible in this shot. IF a person appears, they must be one of: 습격당한 남자 (한국인 성인 남성, 짧은 검은 머리, 자연스러운 얼굴형); 정체불명의 아시아인 남성 (포식 활성 상태) (동아시아계 성인 남성, 짧은 검은 머리, 자연스러운 남성 얼굴) — never anyone else, and never add a person the shot text does not show. IF only part of a person is in frame (a hand, arm, foot, back, silhouette), that body part belongs to the specific person the shot text names — its sex, age, build, skin and grooming must unmistakably match that person's profile above.

No text, captions, watermarks or annotations anywhere.

VISUAL APPROACH — this take (camera, framing, composition):
Use an immersive wide-angle feel with deep focus, keeping the enclosing woodland and both complete bodies sharply legible at the instant of impact. Let cool, soft dusk ambience model the figures naturally, with restrained edge separation preserving readable human features on the dark incoming man. Freeze the violent shoulder contact and collapsing body mechanics with crisp detail, while the slight cant and converging undergrowth intensify the loss of balance.

PRIORITY OF INSTRUCTIONS: the VISUAL APPROACH above governs camera,
framing and composition ONLY. Every other instruction in this prompt —
location, time of day, people and their identity, poses, props,
carried state, physical rules, and the no-text rule — is binding
exactly as written earlier. If the VISUAL APPROACH appears to conflict
with any of them, the earlier instruction wins.

Create ONE FINAL photorealistic live-action film still of the moment below — contemporary South Korea, 2026; all people are Korean unless stated. TIME OF DAY (lock): dusk.

SHOT TEXT (authoritative, Korean): 어두운 실루엣의 낯선 아시아인 남성이 숲속에서 습격당한 남자의 어깨를 강하게 덮치며 함께 바닥으로 쓰러지는 mid-impact 전신

LOCATION (lock): In a patch of dense woodland undergrowth at dusk, where the fleeing man is knocked to the forest floor. The shot takes place here — the attached PREVIOUS SHOT STILL shows this exact place.

CAMERA & FRAME (follow exactly — this is the composition authority for this still; the reference images supply identity and place, never the framing):
- CAMERA: At waist height and several meters away, the camera reaches the end of its rapid dolly-out in a wide, slightly canted lateral view, preserving both figures head to foot as their combined momentum drops toward the lower center. The fleeing man occupies middle-right while the dark incoming figure cuts across from middle-left, with the forest retained around their fall for scale.
- FRAMING SCALE: wide shot
- FRAME LAYOUT: 낯선 아시아인 남성 in the middle-left of the frame, midground, moves toward ground at lower center; 습격당한 남자 in the middle-right of the frame, midground, moves toward ground at lower center; impact ground in the lower-center of the frame, background.
- KEY BACKGROUND ELEMENTS: 우거진 수풀 (Dense forest growth); used as Encloses the collision and supplies full-body scale around the falling figures; 숲 바닥 (State not otherwise specified); used as Landing plane beneath the two falling bodies.
Compose the frame exactly as specified above — camera angle, subject scale and screen placement. Keep true physical scale between people and background elements: fixtures and distant objects occupy only the small screen area their real size and distance dictate; never enlarge a background object into a foreground presence.

THIS SHOT CONTINUES THE PREVIOUS SHOT: everything the attached still established — the place, its fixed features and wear, each person's clothing and state — persists. Any person in it who cannot move stays PRECISELY as photographed (body, pose, contact points, held objects); only the camera changes.

REALIZE FIGURATIVE LANGUAGE AS A LIVE-ACTION SHOT: the Korean shot
text may describe characters metaphorically, figuratively or with
exaggeration. Photograph what a real movie camera would actually
record on a physical set — exaggerated or figurative impressions
become realistic staging choices (lighting, distance, angle,
wardrobe), not literal fantasy imagery.
EVERY CHARACTER IS A HUMAN BEING: unless the story explicitly
features non-human or virtual beings (as in science fiction or
fantasy), every character — however indirectly, vaguely or
figuratively the text describes them — IS a real human. When the
text gives no direct visual description of a person, IMAGINE one
and still show them as a concrete, fully-formed human being:
always render the human form to the maximum extent — build,
posture, face, hands, clothing — never reduce a person to a shape,
blob, solid silhouette or abstract mass.

EXPRESSIONS ARE ACTED, NEVER ANATOMICAL: when the text describes a
person's eyes, face or presence emotionally or figuratively (vacant,
hollow, dazed, burning, lifeless gaze and the like), it describes an
ACTOR'S PERFORMANCE captured by a real camera — realize it ONLY
through gaze direction, focus, eyelids, facial muscles, stillness
and posture. Every living person's eyes remain anatomically normal
human eyes with a natural iris and pupil, natural sclera and normal
proportions; NEVER whiten, blank out, cloud over, glow, enlarge or
otherwise alter eyeballs, skin or anatomy — unless the story
explicitly declares that being non-human or supernatural in form.

PROPS FACE THE RIGHT WAY: every handheld or used object must be
oriented exactly as its real-world use requires. A person reading,
watching or operating something (a phone, a photograph, a paper,
any device) has its functional side — screen, front, page — facing
THEIR OWN eyes; the camera then sees whatever side the staging
geometry implies (often its back). Show the functional side to the
camera ONLY when the shot text itself stages it toward the viewer.
Never flip, mirror or reverse an object's front and back.

CARRIED STATE (persist exactly — must match the neighbouring shots of this scene): As he is knocked to the ground, the man still has one shoe missing and the bleeding bite mark on his thigh.

PEOPLE: the SHOT TEXT alone decides whether any person is visible in this shot. IF a person appears, they must be one of: 습격당한 남자 (한국인 성인 남성, 짧은 검은 머리, 자연스러운 얼굴형); 정체불명의 아시아인 남성 (포식 활성 상태) (동아시아계 성인 남성, 짧은 검은 머리, 자연스러운 남성 얼굴) — never anyone else, and never add a person the shot text does not show. IF only part of a person is in frame (a hand, arm, foot, back, silhouette), that body part belongs to the specific person the shot text names — its sex, age, build, skin and grooming must unmistakably match that person's profile above.

No text, captions, watermarks or annotations anywhere.

VISUAL APPROACH — this take (camera, framing, composition):
Use a mildly compressed lens feel that stacks the dense growth behind the falling figures, making the woodland seem to close around the collision while retaining the prescribed wide full-body frame. Render the dusk light low-contrast and dim, with faces and clothing still physically readable rather than reduced to featureless silhouettes. Allow controlled directional motion blur along the attacker’s drive and the victim’s downward fall, but hold the shoulder contact and central body mass sufficiently sharp to anchor the mid-impact instant.

PRIORITY OF INSTRUCTIONS: the VISUAL APPROACH above governs camera,
framing and composition ONLY. Every other instruction in this prompt —
location, time of day, people and their identity, poses, props,
carried state, physical rules, and the no-text rule — is binding
exactly as written earlier. If the VISUAL APPROACH appears to conflict
with any of them, the earlier instruction wins.

Edit this photograph. Apply ONLY the corrections listed below,
each in place, changing nothing else:

CORRECTIONS:
- Add the dark fingerless gloves to the attacking man's hands to match his character reference.
- Add blood to the bite mark on the attacked man's thigh so it appears bleeding.

PRESERVE EVERYTHING ELSE EXACTLY as in the original photograph —
every person, pose, face, garment, object, set feature, lighting,
camera and framing stay identical. No text, captions or
watermarks anywhere.

Create ONE FINAL photorealistic live-action film still of the moment below — contemporary South Korea, 2026; all people are Korean unless stated. TIME OF DAY (lock): day.

SHOT TEXT (authoritative, Korean): 강민숙의 손목에 새겨진 작은 원 모양의 뚜렷한 표식 클로즈업

LOCATION (lock): Inside the rooftop apartment near the entryway, with the mother's marked wrist visible in daylight filtering from the small living-room window. The shot takes place here — the attached PREVIOUS SHOT STILL shows this exact place.

CAMERA & FRAME (follow exactly — this is the composition authority for this still; the reference images supply identity and place, never the framing):
- CAMERA: At wrist height, the camera completes its descent along 강민숙's arm and holds static on an isolated detail of the clearly defined circular mark. The wrist crosses the center on a slight diagonal, with only a restrained portion of her hand and sleeve retained as anatomical context.
- FRAMING SCALE: insert close-up on a detail
Compose the frame exactly as specified above — camera angle, subject scale and screen placement. Keep true physical scale between people and background elements: fixtures and distant objects occupy only the small screen area their real size and distance dictate; never enlarge a background object into a foreground presence.

THIS SHOT CONTINUES THE PREVIOUS SHOT: everything the attached still established — the place, its fixed features and wear, each person's clothing and state — persists. Any person in it who cannot move stays PRECISELY as photographed (body, pose, contact points, held objects); only the camera changes.

PREVIOUS STILL USAGE (follow exactly — what to take from the attached still and what to exclude): Take the established rooftop-room interior look, daylight, materials, and the continuing clothing of the two women. Exclude the younger woman’s smiling face and shoulder-embrace composition; frame only the marked wrist and nearby interior context.

REALIZE FIGURATIVE LANGUAGE AS A LIVE-ACTION SHOT: the Korean shot
text may describe characters metaphorically, figuratively or with
exaggeration. Photograph what a real movie camera would actually
record on a physical set — exaggerated or figurative impressions
become realistic staging choices (lighting, distance, angle,
wardrobe), not literal fantasy imagery.
EVERY CHARACTER IS A HUMAN BEING: unless the story explicitly
features non-human or virtual beings (as in science fiction or
fantasy), every character — however indirectly, vaguely or
figuratively the text describes them — IS a real human. When the
text gives no direct visual description of a person, IMAGINE one
and still show them as a concrete, fully-formed human being:
always render the human form to the maximum extent — build,
posture, face, hands, clothing — never reduce a person to a shape,
blob, solid silhouette or abstract mass.

EXPRESSIONS ARE ACTED, NEVER ANATOMICAL: when the text describes a
person's eyes, face or presence emotionally or figuratively (vacant,
hollow, dazed, burning, lifeless gaze and the like), it describes an
ACTOR'S PERFORMANCE captured by a real camera — realize it ONLY
through gaze direction, focus, eyelids, facial muscles, stillness
and posture. Every living person's eyes remain anatomically normal
human eyes with a natural iris and pupil, natural sclera and normal
proportions; NEVER whiten, blank out, cloud over, glow, enlarge or
otherwise alter eyeballs, skin or anatomy — unless the story
explicitly declares that being non-human or supernatural in form.

PROPS FACE THE RIGHT WAY: every handheld or used object must be
oriented exactly as its real-world use requires. A person reading,
watching or operating something (a phone, a photograph, a paper,
any device) has its functional side — screen, front, page — facing
THEIR OWN eyes; the camera then sees whatever side the staging
geometry implies (often its back). Show the functional side to the
camera ONLY when the shot text itself stages it toward the viewer.
Never flip, mirror or reverse an object's front and back.

CARRIED STATE (persist exactly — must match the neighbouring shots of this scene): The same small circular mark that Min-suk has had since Su-ri-young’s childhood is clearly present on her wrist.

PEOPLE: the SHOT TEXT alone decides whether any person is visible in this shot. IF a person appears, they must be one of: 강민숙 (한국인 40대 초반 여성, 단정한 검은 머리, 자연스러운 중년 얼굴) — never anyone else, and never add a person the shot text does not show. IF only part of a person is in frame (a hand, arm, foot, back, silhouette), that body part belongs to the specific person the shot text names — its sex, age, build, skin and grooming must unmistakably match that person's profile above.

No text, captions, watermarks or annotations anywhere.

Create ONE FINAL photorealistic live-action film still of the moment below — contemporary South Korea, 2026; all people are Korean unless stated. TIME OF DAY (lock): day.

SHOT TEXT (authoritative, Korean): 강민숙의 손목에 새겨진 작은 원 모양의 뚜렷한 표식 클로즈업

LOCATION (lock): Inside the rooftop apartment near the entryway, with the mother's marked wrist visible in daylight filtering from the small living-room window. The shot takes place here — the attached PREVIOUS SHOT STILL shows this exact place.

CAMERA & FRAME (follow exactly — this is the composition authority for this still; the reference images supply identity and place, never the framing):
- CAMERA: At wrist height, the camera completes its descent along 강민숙's arm and holds static on an isolated detail of the clearly defined circular mark. The wrist crosses the center on a slight diagonal, with only a restrained portion of her hand and sleeve retained as anatomical context.
- FRAMING SCALE: insert close-up on a detail
Compose the frame exactly as specified above — camera angle, subject scale and screen placement. Keep true physical scale between people and background elements: fixtures and distant objects occupy only the small screen area their real size and distance dictate; never enlarge a background object into a foreground presence.

THIS SHOT CONTINUES THE PREVIOUS SHOT: everything the attached still established — the place, its fixed features and wear, each person's clothing and state — persists. Any person in it who cannot move stays PRECISELY as photographed (body, pose, contact points, held objects); only the camera changes.

PREVIOUS STILL USAGE (follow exactly — what to take from the attached still and what to exclude): Take the established rooftop-room interior look, daylight, materials, and the continuing clothing of the two women. Exclude the younger woman’s smiling face and shoulder-embrace composition; frame only the marked wrist and nearby interior context.

REALIZE FIGURATIVE LANGUAGE AS A LIVE-ACTION SHOT: the Korean shot
text may describe characters metaphorically, figuratively or with
exaggeration. Photograph what a real movie camera would actually
record on a physical set — exaggerated or figurative impressions
become realistic staging choices (lighting, distance, angle,
wardrobe), not literal fantasy imagery.
EVERY CHARACTER IS A HUMAN BEING: unless the story explicitly
features non-human or virtual beings (as in science fiction or
fantasy), every character — however indirectly, vaguely or
figuratively the text describes them — IS a real human. When the
text gives no direct visual description of a person, IMAGINE one
and still show them as a concrete, fully-formed human being:
always render the human form to the maximum extent — build,
posture, face, hands, clothing — never reduce a person to a shape,
blob, solid silhouette or abstract mass.

EXPRESSIONS ARE ACTED, NEVER ANATOMICAL: when the text describes a
person's eyes, face or presence emotionally or figuratively (vacant,
hollow, dazed, burning, lifeless gaze and the like), it describes an
ACTOR'S PERFORMANCE captured by a real camera — realize it ONLY
through gaze direction, focus, eyelids, facial muscles, stillness
and posture. Every living person's eyes remain anatomically normal
human eyes with a natural iris and pupil, natural sclera and normal
proportions; NEVER whiten, blank out, cloud over, glow, enlarge or
otherwise alter eyeballs, skin or anatomy — unless the story
explicitly declares that being non-human or supernatural in form.

PROPS FACE THE RIGHT WAY: every handheld or used object must be
oriented exactly as its real-world use requires. A person reading,
watching or operating something (a phone, a photograph, a paper,
any device) has its functional side — screen, front, page — facing
THEIR OWN eyes; the camera then sees whatever side the staging
geometry implies (often its back). Show the functional side to the
camera ONLY when the shot text itself stages it toward the viewer.
Never flip, mirror or reverse an object's front and back.

CARRIED STATE (persist exactly — must match the neighbouring shots of this scene): The same small circular mark that Min-suk has had since Su-ri-young’s childhood is clearly present on her wrist.

PEOPLE: the SHOT TEXT alone decides whether any person is visible in this shot. IF a person appears, they must be one of: 강민숙 (한국인 40대 초반 여성, 단정한 검은 머리, 자연스러운 중년 얼굴) — never anyone else, and never add a person the shot text does not show. IF only part of a person is in frame (a hand, arm, foot, back, silhouette), that body part belongs to the specific person the shot text names — its sex, age, build, skin and grooming must unmistakably match that person's profile above.

No text, captions, watermarks or annotations anywhere.

VISUAL APPROACH — this take (camera, framing, composition):
Use a close, nearly orthographic lens feel with minimal perspective distortion, holding the wrist as a clean graphic diagonal through the center. Keep the circular mark and surrounding natural skin texture crisply resolved, with restrained depth falloff and soft, even window daylight; let the nearby interior dissolve into quiet, unobtrusive tonal context.

PRIORITY OF INSTRUCTIONS: the VISUAL APPROACH above governs camera,
framing and composition ONLY. Every other instruction in this prompt —
location, time of day, people and their identity, poses, props,
carried state, physical rules, and the no-text rule — is binding
exactly as written earlier. If the VISUAL APPROACH appears to conflict
with any of them, the earlier instruction wins.

Create ONE FINAL photorealistic live-action film still of the moment below — contemporary South Korea, 2026; all people are Korean unless stated. TIME OF DAY (lock): day.

SHOT TEXT (authoritative, Korean): 강민숙의 손목에 새겨진 작은 원 모양의 뚜렷한 표식 클로즈업

LOCATION (lock): Inside the rooftop apartment near the entryway, with the mother's marked wrist visible in daylight filtering from the small living-room window. The shot takes place here — the attached PREVIOUS SHOT STILL shows this exact place.

CAMERA & FRAME (follow exactly — this is the composition authority for this still; the reference images supply identity and place, never the framing):
- CAMERA: At wrist height, the camera completes its descent along 강민숙's arm and holds static on an isolated detail of the clearly defined circular mark. The wrist crosses the center on a slight diagonal, with only a restrained portion of her hand and sleeve retained as anatomical context.
- FRAMING SCALE: insert close-up on a detail
Compose the frame exactly as specified above — camera angle, subject scale and screen placement. Keep true physical scale between people and background elements: fixtures and distant objects occupy only the small screen area their real size and distance dictate; never enlarge a background object into a foreground presence.

THIS SHOT CONTINUES THE PREVIOUS SHOT: everything the attached still established — the place, its fixed features and wear, each person's clothing and state — persists. Any person in it who cannot move stays PRECISELY as photographed (body, pose, contact points, held objects); only the camera changes.

PREVIOUS STILL USAGE (follow exactly — what to take from the attached still and what to exclude): Take the established rooftop-room interior look, daylight, materials, and the continuing clothing of the two women. Exclude the younger woman’s smiling face and shoulder-embrace composition; frame only the marked wrist and nearby interior context.

REALIZE FIGURATIVE LANGUAGE AS A LIVE-ACTION SHOT: the Korean shot
text may describe characters metaphorically, figuratively or with
exaggeration. Photograph what a real movie camera would actually
record on a physical set — exaggerated or figurative impressions
become realistic staging choices (lighting, distance, angle,
wardrobe), not literal fantasy imagery.
EVERY CHARACTER IS A HUMAN BEING: unless the story explicitly
features non-human or virtual beings (as in science fiction or
fantasy), every character — however indirectly, vaguely or
figuratively the text describes them — IS a real human. When the
text gives no direct visual description of a person, IMAGINE one
and still show them as a concrete, fully-formed human being:
always render the human form to the maximum extent — build,
posture, face, hands, clothing — never reduce a person to a shape,
blob, solid silhouette or abstract mass.

EXPRESSIONS ARE ACTED, NEVER ANATOMICAL: when the text describes a
person's eyes, face or presence emotionally or figuratively (vacant,
hollow, dazed, burning, lifeless gaze and the like), it describes an
ACTOR'S PERFORMANCE captured by a real camera — realize it ONLY
through gaze direction, focus, eyelids, facial muscles, stillness
and posture. Every living person's eyes remain anatomically normal
human eyes with a natural iris and pupil, natural sclera and normal
proportions; NEVER whiten, blank out, cloud over, glow, enlarge or
otherwise alter eyeballs, skin or anatomy — unless the story
explicitly declares that being non-human or supernatural in form.

PROPS FACE THE RIGHT WAY: every handheld or used object must be
oriented exactly as its real-world use requires. A person reading,
watching or operating something (a phone, a photograph, a paper,
any device) has its functional side — screen, front, page — facing
THEIR OWN eyes; the camera then sees whatever side the staging
geometry implies (often its back). Show the functional side to the
camera ONLY when the shot text itself stages it toward the viewer.
Never flip, mirror or reverse an object's front and back.

CARRIED STATE (persist exactly — must match the neighbouring shots of this scene): The same small circular mark that Min-suk has had since Su-ri-young’s childhood is clearly present on her wrist.

PEOPLE: the SHOT TEXT alone decides whether any person is visible in this shot. IF a person appears, they must be one of: 강민숙 (한국인 40대 초반 여성, 단정한 검은 머리, 자연스러운 중년 얼굴) — never anyone else, and never add a person the shot text does not show. IF only part of a person is in frame (a hand, arm, foot, back, silhouette), that body part belongs to the specific person the shot text names — its sex, age, build, skin and grooming must unmistakably match that person's profile above.

No text, captions, watermarks or annotations anywhere.

VISUAL APPROACH — this take (camera, framing, composition):
Photograph the detail with an intimate macro-like perspective and very shallow focus, placing the sharpest plane precisely on the circular mark while the retained hand, sleeve, and nearby room fall away progressively. Let the window daylight graze across the wrist to reveal subtle skin relief and give the mark tactile prominence, with subdued background depth and gentle natural contrast.

PRIORITY OF INSTRUCTIONS: the VISUAL APPROACH above governs camera,
framing and composition ONLY. Every other instruction in this prompt —
location, time of day, people and their identity, poses, props,
carried state, physical rules, and the no-text rule — is binding
exactly as written earlier. If the VISUAL APPROACH appears to conflict
with any of them, the earlier instruction wins.

Edit this photograph. Apply ONLY the corrections listed below,
each in place, changing nothing else:

CORRECTIONS:
- Rotate the wrist so the outer/side aspect faces the camera, and change the red scar back to the thin, dark circular outline established in the previous still.

PRESERVE EVERYTHING ELSE EXACTLY as in the original photograph —
every person, pose, face, garment, object, set feature, lighting,
camera and framing stay identical. No text, captions or
watermarks anywhere.

Create ONE FINAL photorealistic live-action film still of the moment below — contemporary South Korea, 2026; all people are Korean unless stated. TIME OF DAY (lock): day.

SHOT TEXT (authoritative, Korean): 철창에 갇힌 개들 앞에서 수리영이 휴대폰을 귀에 댄 채 굳은 표정으로 서 있는 상체

LOCATION (lock): Along the shelter's kennel row, in front of the enclosures holding the dogs. The shot takes place here — the attached PREVIOUS SHOT STILL shows this exact place.

CAMERA & FRAME (follow exactly — this is the composition authority for this still; the reference images supply identity and place, never the framing):
- CAMERA: At chest height in front of 수리영, the lateral track settles into a brief medium hold, placing her rigid upper body in the left-center with the phone clearly pressed to her ear. The row of occupied enclosures recedes across the right background, preserving the concerned pause before the camera resumes tracking toward her work area.
- FRAMING SCALE: medium shot
- FRAME LAYOUT: 수리영 in the middle-left of the frame, midground; row of enclosures with confined dogs in the middle-right of the frame, background.
- KEY BACKGROUND ELEMENTS: 철창 우리 (Closed with dogs confined inside) — The barred fronts face the camera and continue into the background beside her; used as Receding background structure that keeps the confined dogs visible behind 수리영; 휴대폰 (Used for an unanswered call) — Its screen faces away from the camera while one side rests against her ear; used as Primary handheld prop kept readable beside her fixed expression.
Compose the frame exactly as specified above — camera angle, subject scale and screen placement. Keep true physical scale between people and background elements: fixtures and distant objects occupy only the small screen area their real size and distance dictate; never enlarge a background object into a foreground presence.

THIS SHOT CONTINUES THE PREVIOUS SHOT: everything the attached still established — the place, its fixed features and wear, each person's clothing and state — persists. Any person in it who cannot move stays PRECISELY as photographed (body, pose, contact points, held objects); only the camera changes.

REALIZE FIGURATIVE LANGUAGE AS A LIVE-ACTION SHOT: the Korean shot
text may describe characters metaphorically, figuratively or with
exaggeration. Photograph what a real movie camera would actually
record on a physical set — exaggerated or figurative impressions
become realistic staging choices (lighting, distance, angle,
wardrobe), not literal fantasy imagery.
EVERY CHARACTER IS A HUMAN BEING: unless the story explicitly
features non-human or virtual beings (as in science fiction or
fantasy), every character — however indirectly, vaguely or
figuratively the text describes them — IS a real human. When the
text gives no direct visual description of a person, IMAGINE one
and still show them as a concrete, fully-formed human being:
always render the human form to the maximum extent — build,
posture, face, hands, clothing — never reduce a person to a shape,
blob, solid silhouette or abstract mass.

EXPRESSIONS ARE ACTED, NEVER ANATOMICAL: when the text describes a
person's eyes, face or presence emotionally or figuratively (vacant,
hollow, dazed, burning, lifeless gaze and the like), it describes an
ACTOR'S PERFORMANCE captured by a real camera — realize it ONLY
through gaze direction, focus, eyelids, facial muscles, stillness
and posture. Every living person's eyes remain anatomically normal
human eyes with a natural iris and pupil, natural sclera and normal
proportions; NEVER whiten, blank out, cloud over, glow, enlarge or
otherwise alter eyeballs, skin or anatomy — unless the story
explicitly declares that being non-human or supernatural in form.

PROPS FACE THE RIGHT WAY: every handheld or used object must be
oriented exactly as its real-world use requires. A person reading,
watching or operating something (a phone, a photograph, a paper,
any device) has its functional side — screen, front, page — facing
THEIR OWN eyes; the camera then sees whatever side the staging
geometry implies (often its back). Show the functional side to the
camera ONLY when the shot text itself stages it toward the viewer.
Never flip, mirror or reverse an object's front and back.

CARRIED STATE (persist exactly — must match the neighbouring shots of this scene): Su-ri-young has her phone with her while the shelter dogs remain confined in their cages.

PEOPLE: the SHOT TEXT alone decides whether any person is visible in this shot. IF a person appears, they must be one of: 수리영 (한국인 20대 초반 여성, 긴 검은 생머리, 젊은 타원형 얼굴); 로트와일러 (검정·갈색 털, 넓은 머리, 짧고 굵은 주둥이, 접힌 삼각형 귀) — never anyone else, and never add a person the shot text does not show. IF only part of a person is in frame (a hand, arm, foot, back, silhouette), that body part belongs to the specific person the shot text names — its sex, age, build, skin and grooming must unmistakably match that person's profile above.

No text, captions, watermarks or annotations anywhere.

Create ONE FINAL photorealistic live-action film still of the moment below — contemporary South Korea, 2026; all people are Korean unless stated. TIME OF DAY (lock): day.

SHOT TEXT (authoritative, Korean): 철창에 갇힌 개들 앞에서 수리영이 휴대폰을 귀에 댄 채 굳은 표정으로 서 있는 상체

LOCATION (lock): Along the shelter's kennel row, in front of the enclosures holding the dogs. The shot takes place here — the attached PREVIOUS SHOT STILL shows this exact place.

CAMERA & FRAME (follow exactly — this is the composition authority for this still; the reference images supply identity and place, never the framing):
- CAMERA: At chest height in front of 수리영, the lateral track settles into a brief medium hold, placing her rigid upper body in the left-center with the phone clearly pressed to her ear. The row of occupied enclosures recedes across the right background, preserving the concerned pause before the camera resumes tracking toward her work area.
- FRAMING SCALE: medium shot
- FRAME LAYOUT: 수리영 in the middle-left of the frame, midground; row of enclosures with confined dogs in the middle-right of the frame, background.
- KEY BACKGROUND ELEMENTS: 철창 우리 (Closed with dogs confined inside) — The barred fronts face the camera and continue into the background beside her; used as Receding background structure that keeps the confined dogs visible behind 수리영; 휴대폰 (Used for an unanswered call) — Its screen faces away from the camera while one side rests against her ear; used as Primary handheld prop kept readable beside her fixed expression.
Compose the frame exactly as specified above — camera angle, subject scale and screen placement. Keep true physical scale between people and background elements: fixtures and distant objects occupy only the small screen area their real size and distance dictate; never enlarge a background object into a foreground presence.

THIS SHOT CONTINUES THE PREVIOUS SHOT: everything the attached still established — the place, its fixed features and wear, each person's clothing and state — persists. Any person in it who cannot move stays PRECISELY as photographed (body, pose, contact points, held objects); only the camera changes.

REALIZE FIGURATIVE LANGUAGE AS A LIVE-ACTION SHOT: the Korean shot
text may describe characters metaphorically, figuratively or with
exaggeration. Photograph what a real movie camera would actually
record on a physical set — exaggerated or figurative impressions
become realistic staging choices (lighting, distance, angle,
wardrobe), not literal fantasy imagery.
EVERY CHARACTER IS A HUMAN BEING: unless the story explicitly
features non-human or virtual beings (as in science fiction or
fantasy), every character — however indirectly, vaguely or
figuratively the text describes them — IS a real human. When the
text gives no direct visual description of a person, IMAGINE one
and still show them as a concrete, fully-formed human being:
always render the human form to the maximum extent — build,
posture, face, hands, clothing — never reduce a person to a shape,
blob, solid silhouette or abstract mass.

EXPRESSIONS ARE ACTED, NEVER ANATOMICAL: when the text describes a
person's eyes, face or presence emotionally or figuratively (vacant,
hollow, dazed, burning, lifeless gaze and the like), it describes an
ACTOR'S PERFORMANCE captured by a real camera — realize it ONLY
through gaze direction, focus, eyelids, facial muscles, stillness
and posture. Every living person's eyes remain anatomically normal
human eyes with a natural iris and pupil, natural sclera and normal
proportions; NEVER whiten, blank out, cloud over, glow, enlarge or
otherwise alter eyeballs, skin or anatomy — unless the story
explicitly declares that being non-human or supernatural in form.

PROPS FACE THE RIGHT WAY: every handheld or used object must be
oriented exactly as its real-world use requires. A person reading,
watching or operating something (a phone, a photograph, a paper,
any device) has its functional side — screen, front, page — facing
THEIR OWN eyes; the camera then sees whatever side the staging
geometry implies (often its back). Show the functional side to the
camera ONLY when the shot text itself stages it toward the viewer.
Never flip, mirror or reverse an object's front and back.

CARRIED STATE (persist exactly — must match the neighbouring shots of this scene): Su-ri-young has her phone with her while the shelter dogs remain confined in their cages.

PEOPLE: the SHOT TEXT alone decides whether any person is visible in this shot. IF a person appears, they must be one of: 수리영 (한국인 20대 초반 여성, 긴 검은 생머리, 젊은 타원형 얼굴); 로트와일러 (검정·갈색 털, 넓은 머리, 짧고 굵은 주둥이, 접힌 삼각형 귀) — never anyone else, and never add a person the shot text does not show. IF only part of a person is in frame (a hand, arm, foot, back, silhouette), that body part belongs to the specific person the shot text names — its sex, age, build, skin and grooming must unmistakably match that person's profile above.

No text, captions, watermarks or annotations anywhere.

VISUAL APPROACH — this take (camera, framing, composition):
Use a restrained telephoto feel with compressed perspective and shallow depth of field, holding Su-ri-young’s face, rigid posture, and phone in crisp focus while the barred kennel row and confined dogs remain recognizable but gently softened behind her. Shape the available daytime light softly across her face, with modest contrast and subdued color, turning the pause into an intimate, suspended moment.

PRIORITY OF INSTRUCTIONS: the VISUAL APPROACH above governs camera,
framing and composition ONLY. Every other instruction in this prompt —
location, time of day, people and their identity, poses, props,
carried state, physical rules, and the no-text rule — is binding
exactly as written earlier. If the VISUAL APPROACH appears to conflict
with any of them, the earlier instruction wins.

Create ONE FINAL photorealistic live-action film still of the moment below — contemporary South Korea, 2026; all people are Korean unless stated. TIME OF DAY (lock): day.

SHOT TEXT (authoritative, Korean): 철창에 갇힌 개들 앞에서 수리영이 휴대폰을 귀에 댄 채 굳은 표정으로 서 있는 상체

LOCATION (lock): Along the shelter's kennel row, in front of the enclosures holding the dogs. The shot takes place here — the attached PREVIOUS SHOT STILL shows this exact place.

CAMERA & FRAME (follow exactly — this is the composition authority for this still; the reference images supply identity and place, never the framing):
- CAMERA: At chest height in front of 수리영, the lateral track settles into a brief medium hold, placing her rigid upper body in the left-center with the phone clearly pressed to her ear. The row of occupied enclosures recedes across the right background, preserving the concerned pause before the camera resumes tracking toward her work area.
- FRAMING SCALE: medium shot
- FRAME LAYOUT: 수리영 in the middle-left of the frame, midground; row of enclosures with confined dogs in the middle-right of the frame, background.
- KEY BACKGROUND ELEMENTS: 철창 우리 (Closed with dogs confined inside) — The barred fronts face the camera and continue into the background beside her; used as Receding background structure that keeps the confined dogs visible behind 수리영; 휴대폰 (Used for an unanswered call) — Its screen faces away from the camera while one side rests against her ear; used as Primary handheld prop kept readable beside her fixed expression.
Compose the frame exactly as specified above — camera angle, subject scale and screen placement. Keep true physical scale between people and background elements: fixtures and distant objects occupy only the small screen area their real size and distance dictate; never enlarge a background object into a foreground presence.

THIS SHOT CONTINUES THE PREVIOUS SHOT: everything the attached still established — the place, its fixed features and wear, each person's clothing and state — persists. Any person in it who cannot move stays PRECISELY as photographed (body, pose, contact points, held objects); only the camera changes.

REALIZE FIGURATIVE LANGUAGE AS A LIVE-ACTION SHOT: the Korean shot
text may describe characters metaphorically, figuratively or with
exaggeration. Photograph what a real movie camera would actually
record on a physical set — exaggerated or figurative impressions
become realistic staging choices (lighting, distance, angle,
wardrobe), not literal fantasy imagery.
EVERY CHARACTER IS A HUMAN BEING: unless the story explicitly
features non-human or virtual beings (as in science fiction or
fantasy), every character — however indirectly, vaguely or
figuratively the text describes them — IS a real human. When the
text gives no direct visual description of a person, IMAGINE one
and still show them as a concrete, fully-formed human being:
always render the human form to the maximum extent — build,
posture, face, hands, clothing — never reduce a person to a shape,
blob, solid silhouette or abstract mass.

EXPRESSIONS ARE ACTED, NEVER ANATOMICAL: when the text describes a
person's eyes, face or presence emotionally or figuratively (vacant,
hollow, dazed, burning, lifeless gaze and the like), it describes an
ACTOR'S PERFORMANCE captured by a real camera — realize it ONLY
through gaze direction, focus, eyelids, facial muscles, stillness
and posture. Every living person's eyes remain anatomically normal
human eyes with a natural iris and pupil, natural sclera and normal
proportions; NEVER whiten, blank out, cloud over, glow, enlarge or
otherwise alter eyeballs, skin or anatomy — unless the story
explicitly declares that being non-human or supernatural in form.

PROPS FACE THE RIGHT WAY: every handheld or used object must be
oriented exactly as its real-world use requires. A person reading,
watching or operating something (a phone, a photograph, a paper,
any device) has its functional side — screen, front, page — facing
THEIR OWN eyes; the camera then sees whatever side the staging
geometry implies (often its back). Show the functional side to the
camera ONLY when the shot text itself stages it toward the viewer.
Never flip, mirror or reverse an object's front and back.

CARRIED STATE (persist exactly — must match the neighbouring shots of this scene): Su-ri-young has her phone with her while the shelter dogs remain confined in their cages.

PEOPLE: the SHOT TEXT alone decides whether any person is visible in this shot. IF a person appears, they must be one of: 수리영 (한국인 20대 초반 여성, 긴 검은 생머리, 젊은 타원형 얼굴); 로트와일러 (검정·갈색 털, 넓은 머리, 짧고 굵은 주둥이, 접힌 삼각형 귀) — never anyone else, and never add a person the shot text does not show. IF only part of a person is in frame (a hand, arm, foot, back, silhouette), that body part belongs to the specific person the shot text names — its sex, age, build, skin and grooming must unmistakably match that person's profile above.

No text, captions, watermarks or annotations anywhere.

VISUAL APPROACH — this take (camera, framing, composition):
Use a natural wide-lens feel without distorting the medium framing, preserving deep focus so Su-ri-young, the repeated bars, and the confined dogs share the same sharply legible space. Emphasize the kennel row’s receding lines across the right side as a visual pressure against her stillness, with harder directional daytime illumination revealing the enclosure texture and a cool, observational realism.

PRIORITY OF INSTRUCTIONS: the VISUAL APPROACH above governs camera,
framing and composition ONLY. Every other instruction in this prompt —
location, time of day, people and their identity, poses, props,
carried state, physical rules, and the no-text rule — is binding
exactly as written earlier. If the VISUAL APPROACH appears to conflict
with any of them, the earlier instruction wins.

Edit this photograph. Apply ONLY the corrections listed below,
each in place, changing nothing else:

CORRECTIONS:
- Rotate the phone so its screen faces away from the camera.
- Remove the backpack strap from her right shoulder to match the reference image's one-shoulder carry.
- Replace the light-colored dogs in the background enclosures with Rottweilers.
- Correct the hand anatomy to show a normal structure with the correct number of fingers.

PRESERVE EVERYTHING ELSE EXACTLY as in the original photograph —
every person, pose, face, garment, object, set feature, lighting,
camera and framing stay identical. No text, captions or
watermarks anywhere.

Create ONE FINAL photorealistic live-action film still of the moment below — contemporary South Korea, 2026; all people are Korean unless stated. TIME OF DAY (lock): night.

SHOT TEXT (authoritative, Korean): 죽은 강민숙의 굳어버린 손안에 구겨진 종이 사진이 쥐어져 있는 클로즈업

LOCATION (lock): On the floor behind the curtain in the daughter's bedroom, beside the dead mother's body in the dim nighttime apartment. The shot takes place here — the attached PREVIOUS SHOT STILL shows this exact place.

CAMERA & FRAME (follow exactly — this is the composition authority for this still; the reference images supply identity and place, never the framing):
- CAMERA: At hand height, a steep but not overhead downward angle isolates 강민숙’s rigid fingers and the crumpled photograph while retaining a narrow strip of her forearm and the nearby floor for scale. The hand lies across the lower-center and the photograph remains below forty percent of the frame, its creased image side caught between the closed fingers.
- FRAMING SCALE: insert close-up on a detail
- KEY BACKGROUND ELEMENTS: 구겨진 종이 사진 (Crumpled and gripped inside the hand) — The camera sees the crumpled image-bearing face, partially concealed by fingers and folds; used as Central clue held within the rigid hand; 바닥 (No additional state specified in this framing); used as Scale reference at the edge of the isolated detail.
Compose the frame exactly as specified above — camera angle, subject scale and screen placement. Keep true physical scale between people and background elements: fixtures and distant objects occupy only the small screen area their real size and distance dictate; never enlarge a background object into a foreground presence.

THIS SHOT CONTINUES THE PREVIOUS SHOT: everything the attached still established — the place, its fixed features and wear, each person's clothing and state — persists. Any person in it who cannot move stays PRECISELY as photographed (body, pose, contact points, held objects); only the camera changes.

REALIZE FIGURATIVE LANGUAGE AS A LIVE-ACTION SHOT: the Korean shot
text may describe characters metaphorically, figuratively or with
exaggeration. Photograph what a real movie camera would actually
record on a physical set — exaggerated or figurative impressions
become realistic staging choices (lighting, distance, angle,
wardrobe), not literal fantasy imagery.
EVERY CHARACTER IS A HUMAN BEING: unless the story explicitly
features non-human or virtual beings (as in science fiction or
fantasy), every character — however indirectly, vaguely or
figuratively the text describes them — IS a real human. When the
text gives no direct visual description of a person, IMAGINE one
and still show them as a concrete, fully-formed human being:
always render the human form to the maximum extent — build,
posture, face, hands, clothing — never reduce a person to a shape,
blob, solid silhouette or abstract mass.

EXPRESSIONS ARE ACTED, NEVER ANATOMICAL: when the text describes a
person's eyes, face or presence emotionally or figuratively (vacant,
hollow, dazed, burning, lifeless gaze and the like), it describes an
ACTOR'S PERFORMANCE captured by a real camera — realize it ONLY
through gaze direction, focus, eyelids, facial muscles, stillness
and posture. Every living person's eyes remain anatomically normal
human eyes with a natural iris and pupil, natural sclera and normal
proportions; NEVER whiten, blank out, cloud over, glow, enlarge or
otherwise alter eyeballs, skin or anatomy — unless the story
explicitly declares that being non-human or supernatural in form.

PROPS FACE THE RIGHT WAY: every handheld or used object must be
oriented exactly as its real-world use requires. A person reading,
watching or operating something (a phone, a photograph, a paper,
any device) has its functional side — screen, front, page — facing
THEIR OWN eyes; the camera then sees whatever side the staging
geometry implies (often its back). Show the functional side to the
camera ONLY when the shot text itself stages it toward the viewer.
Never flip, mirror or reverse an object's front and back.

CARRIED STATE (persist exactly — must match the neighbouring shots of this scene): Min-suk remains dead. She is seated motionless on the floor behind the curtain with her torso slumped and her head dropped forward; one rigid hand is closed around a crumpled photograph. The circular wrist mark is present and turns red as Su-ri-young removes the photograph.

NO PEOPLE IN THIS SHOT: the shot text shows only the place and its state — no living person or any body part appears in frame, unless the shot text itself explicitly says so.

No text, captions, watermarks or annotations anywhere.

Create ONE FINAL photorealistic live-action film still of the moment below — contemporary South Korea, 2026; all people are Korean unless stated. TIME OF DAY (lock): night.

SHOT TEXT (authoritative, Korean): 죽은 강민숙의 굳어버린 손안에 구겨진 종이 사진이 쥐어져 있는 클로즈업

LOCATION (lock): On the floor behind the curtain in the daughter's bedroom, beside the dead mother's body in the dim nighttime apartment. The shot takes place here — the attached PREVIOUS SHOT STILL shows this exact place.

CAMERA & FRAME (follow exactly — this is the composition authority for this still; the reference images supply identity and place, never the framing):
- CAMERA: At hand height, a steep but not overhead downward angle isolates 강민숙’s rigid fingers and the crumpled photograph while retaining a narrow strip of her forearm and the nearby floor for scale. The hand lies across the lower-center and the photograph remains below forty percent of the frame, its creased image side caught between the closed fingers.
- FRAMING SCALE: insert close-up on a detail
- KEY BACKGROUND ELEMENTS: 구겨진 종이 사진 (Crumpled and gripped inside the hand) — The camera sees the crumpled image-bearing face, partially concealed by fingers and folds; used as Central clue held within the rigid hand; 바닥 (No additional state specified in this framing); used as Scale reference at the edge of the isolated detail.
Compose the frame exactly as specified above — camera angle, subject scale and screen placement. Keep true physical scale between people and background elements: fixtures and distant objects occupy only the small screen area their real size and distance dictate; never enlarge a background object into a foreground presence.

THIS SHOT CONTINUES THE PREVIOUS SHOT: everything the attached still established — the place, its fixed features and wear, each person's clothing and state — persists. Any person in it who cannot move stays PRECISELY as photographed (body, pose, contact points, held objects); only the camera changes.

REALIZE FIGURATIVE LANGUAGE AS A LIVE-ACTION SHOT: the Korean shot
text may describe characters metaphorically, figuratively or with
exaggeration. Photograph what a real movie camera would actually
record on a physical set — exaggerated or figurative impressions
become realistic staging choices (lighting, distance, angle,
wardrobe), not literal fantasy imagery.
EVERY CHARACTER IS A HUMAN BEING: unless the story explicitly
features non-human or virtual beings (as in science fiction or
fantasy), every character — however indirectly, vaguely or
figuratively the text describes them — IS a real human. When the
text gives no direct visual description of a person, IMAGINE one
and still show them as a concrete, fully-formed human being:
always render the human form to the maximum extent — build,
posture, face, hands, clothing — never reduce a person to a shape,
blob, solid silhouette or abstract mass.

EXPRESSIONS ARE ACTED, NEVER ANATOMICAL: when the text describes a
person's eyes, face or presence emotionally or figuratively (vacant,
hollow, dazed, burning, lifeless gaze and the like), it describes an
ACTOR'S PERFORMANCE captured by a real camera — realize it ONLY
through gaze direction, focus, eyelids, facial muscles, stillness
and posture. Every living person's eyes remain anatomically normal
human eyes with a natural iris and pupil, natural sclera and normal
proportions; NEVER whiten, blank out, cloud over, glow, enlarge or
otherwise alter eyeballs, skin or anatomy — unless the story
explicitly declares that being non-human or supernatural in form.

PROPS FACE THE RIGHT WAY: every handheld or used object must be
oriented exactly as its real-world use requires. A person reading,
watching or operating something (a phone, a photograph, a paper,
any device) has its functional side — screen, front, page — facing
THEIR OWN eyes; the camera then sees whatever side the staging
geometry implies (often its back). Show the functional side to the
camera ONLY when the shot text itself stages it toward the viewer.
Never flip, mirror or reverse an object's front and back.

CARRIED STATE (persist exactly — must match the neighbouring shots of this scene): Min-suk remains dead. She is seated motionless on the floor behind the curtain with her torso slumped and her head dropped forward; one rigid hand is closed around a crumpled photograph. The circular wrist mark is present and turns red as Su-ri-young removes the photograph.

NO PEOPLE IN THIS SHOT: the shot text shows only the place and its state — no living person or any body part appears in frame, unless the shot text itself explicitly says so.

No text, captions, watermarks or annotations anywhere.

VISUAL APPROACH — this take (camera, framing, composition):
Use a restrained, forensic visual treatment with crisp focus carried across the rigid fingers, crushed paper folds, and the narrow visible strip of forearm. Let soft, cool nighttime illumination fall obliquely across the hand, creating controlled micro-shadows in the knuckles and creases while the nearby floor recedes quietly; keep the image-bearing face of the photograph legible only to the extent allowed by its concealment and folds.

PRIORITY OF INSTRUCTIONS: the VISUAL APPROACH above governs camera,
framing and composition ONLY. Every other instruction in this prompt —
location, time of day, people and their identity, poses, props,
carried state, physical rules, and the no-text rule — is binding
exactly as written earlier. If the VISUAL APPROACH appears to conflict
with any of them, the earlier instruction wins.

Create ONE FINAL photorealistic live-action film still of the moment below — contemporary South Korea, 2026; all people are Korean unless stated. TIME OF DAY (lock): night.

SHOT TEXT (authoritative, Korean): 죽은 강민숙의 굳어버린 손안에 구겨진 종이 사진이 쥐어져 있는 클로즈업

LOCATION (lock): On the floor behind the curtain in the daughter's bedroom, beside the dead mother's body in the dim nighttime apartment. The shot takes place here — the attached PREVIOUS SHOT STILL shows this exact place.

CAMERA & FRAME (follow exactly — this is the composition authority for this still; the reference images supply identity and place, never the framing):
- CAMERA: At hand height, a steep but not overhead downward angle isolates 강민숙’s rigid fingers and the crumpled photograph while retaining a narrow strip of her forearm and the nearby floor for scale. The hand lies across the lower-center and the photograph remains below forty percent of the frame, its creased image side caught between the closed fingers.
- FRAMING SCALE: insert close-up on a detail
- KEY BACKGROUND ELEMENTS: 구겨진 종이 사진 (Crumpled and gripped inside the hand) — The camera sees the crumpled image-bearing face, partially concealed by fingers and folds; used as Central clue held within the rigid hand; 바닥 (No additional state specified in this framing); used as Scale reference at the edge of the isolated detail.
Compose the frame exactly as specified above — camera angle, subject scale and screen placement. Keep true physical scale between people and background elements: fixtures and distant objects occupy only the small screen area their real size and distance dictate; never enlarge a background object into a foreground presence.

THIS SHOT CONTINUES THE PREVIOUS SHOT: everything the attached still established — the place, its fixed features and wear, each person's clothing and state — persists. Any person in it who cannot move stays PRECISELY as photographed (body, pose, contact points, held objects); only the camera changes.

REALIZE FIGURATIVE LANGUAGE AS A LIVE-ACTION SHOT: the Korean shot
text may describe characters metaphorically, figuratively or with
exaggeration. Photograph what a real movie camera would actually
record on a physical set — exaggerated or figurative impressions
become realistic staging choices (lighting, distance, angle,
wardrobe), not literal fantasy imagery.
EVERY CHARACTER IS A HUMAN BEING: unless the story explicitly
features non-human or virtual beings (as in science fiction or
fantasy), every character — however indirectly, vaguely or
figuratively the text describes them — IS a real human. When the
text gives no direct visual description of a person, IMAGINE one
and still show them as a concrete, fully-formed human being:
always render the human form to the maximum extent — build,
posture, face, hands, clothing — never reduce a person to a shape,
blob, solid silhouette or abstract mass.

EXPRESSIONS ARE ACTED, NEVER ANATOMICAL: when the text describes a
person's eyes, face or presence emotionally or figuratively (vacant,
hollow, dazed, burning, lifeless gaze and the like), it describes an
ACTOR'S PERFORMANCE captured by a real camera — realize it ONLY
through gaze direction, focus, eyelids, facial muscles, stillness
and posture. Every living person's eyes remain anatomically normal
human eyes with a natural iris and pupil, natural sclera and normal
proportions; NEVER whiten, blank out, cloud over, glow, enlarge or
otherwise alter eyeballs, skin or anatomy — unless the story
explicitly declares that being non-human or supernatural in form.

PROPS FACE THE RIGHT WAY: every handheld or used object must be
oriented exactly as its real-world use requires. A person reading,
watching or operating something (a phone, a photograph, a paper,
any device) has its functional side — screen, front, page — facing
THEIR OWN eyes; the camera then sees whatever side the staging
geometry implies (often its back). Show the functional side to the
camera ONLY when the shot text itself stages it toward the viewer.
Never flip, mirror or reverse an object's front and back.

CARRIED STATE (persist exactly — must match the neighbouring shots of this scene): Min-suk remains dead. She is seated motionless on the floor behind the curtain with her torso slumped and her head dropped forward; one rigid hand is closed around a crumpled photograph. The circular wrist mark is present and turns red as Su-ri-young removes the photograph.

NO PEOPLE IN THIS SHOT: the shot text shows only the place and its state — no living person or any body part appears in frame, unless the shot text itself explicitly says so.

No text, captions, watermarks or annotations anywhere.

VISUAL APPROACH — this take (camera, framing, composition):
Render the insert with very shallow, tactile focus concentrated on the pinched edge of the crumpled photograph and the closed fingertips, allowing the nearer forearm edge and surrounding floor to dissolve gently. Use dim raking nighttime light to catch isolated highlights along the paper ridges and rigid finger contours, with restrained shadow falloff making the photograph feel trapped inside the hand.

PRIORITY OF INSTRUCTIONS: the VISUAL APPROACH above governs camera,
framing and composition ONLY. Every other instruction in this prompt —
location, time of day, people and their identity, poses, props,
carried state, physical rules, and the no-text rule — is binding
exactly as written earlier. If the VISUAL APPROACH appears to conflict
with any of them, the earlier instruction wins.

Edit this photograph. Apply ONLY the corrections listed below,
each in place, changing nothing else:

CORRECTIONS:
- Lift the forearm off the floor so it angles downwards with only the hand making contact with the ground.
- Change the camera angle to a steep downward view pointing at the hand and floor.
- Crumple the photograph into a much tighter ball gripped mostly inside the closed fist, matching the reference.

PRESERVE EVERYTHING ELSE EXACTLY as in the original photograph —
every person, pose, face, garment, object, set feature, lighting,
camera and framing stay identical. No text, captions or
watermarks anywhere.

Create ONE FINAL photorealistic live-action film still of the moment below — contemporary South Korea, 2026; all people are Korean unless stated. TIME OF DAY (lock): night.

SHOT TEXT (authoritative, Korean): 침대 위쪽 벽면에 붉은색의 굵은 원 형태가 선명하게 남아 있는 모습

LOCATION (lock): Inside the daughter's bedroom, on the wall directly above the bed in the dim nighttime apartment. The shot takes place here — the attached PREVIOUS SHOT STILL shows this exact place.

CAMERA & FRAME (follow exactly — this is the composition authority for this still; the reference images supply identity and place, never the framing):
- CAMERA: From low bedside height, the upward tilt passes the lower room and settles into a static, slightly low-angle view of the wall above the bed. The complete thick red circle sits in the upper-center at a restrained scale, with the bed edge retained along the lower frame to establish its physical placement rather than turning the mark into an abstract graphic.
- FRAMING SCALE: medium shot
- FRAME LAYOUT: wall with thick red circle above the bed in the upper-center of the frame, background; bed beneath the circle in the lower-center of the frame, midground.
- KEY BACKGROUND ELEMENTS: 침대 위 벽면 (Marked by a clearly visible red circle) — The camera-facing wall plane bears a complete thick red circle above the bed; used as Final supernatural focal surface; 침대 (Present beneath the circle) — Its upper edge is seen below the marked wall from the low bedside angle; used as Lower-frame spatial reference locating the mark.
Compose the frame exactly as specified above — camera angle, subject scale and screen placement. Keep true physical scale between people and background elements: fixtures and distant objects occupy only the small screen area their real size and distance dictate; never enlarge a background object into a foreground presence.

THIS SHOT CONTINUES THE PREVIOUS SHOT: everything the attached still established — the place, its fixed features and wear, each person's clothing and state — persists. Any person in it who cannot move stays PRECISELY as photographed (body, pose, contact points, held objects); only the camera changes.

REALIZE FIGURATIVE LANGUAGE AS A LIVE-ACTION SHOT: the Korean shot
text may describe characters metaphorically, figuratively or with
exaggeration. Photograph what a real movie camera would actually
record on a physical set — exaggerated or figurative impressions
become realistic staging choices (lighting, distance, angle,
wardrobe), not literal fantasy imagery.
EVERY CHARACTER IS A HUMAN BEING: unless the story explicitly
features non-human or virtual beings (as in science fiction or
fantasy), every character — however indirectly, vaguely or
figuratively the text describes them — IS a real human. When the
text gives no direct visual description of a person, IMAGINE one
and still show them as a concrete, fully-formed human being:
always render the human form to the maximum extent — build,
posture, face, hands, clothing — never reduce a person to a shape,
blob, solid silhouette or abstract mass.

EXPRESSIONS ARE ACTED, NEVER ANATOMICAL: when the text describes a
person's eyes, face or presence emotionally or figuratively (vacant,
hollow, dazed, burning, lifeless gaze and the like), it describes an
ACTOR'S PERFORMANCE captured by a real camera — realize it ONLY
through gaze direction, focus, eyelids, facial muscles, stillness
and posture. Every living person's eyes remain anatomically normal
human eyes with a natural iris and pupil, natural sclera and normal
proportions; NEVER whiten, blank out, cloud over, glow, enlarge or
otherwise alter eyeballs, skin or anatomy — unless the story
explicitly declares that being non-human or supernatural in form.

PROPS FACE THE RIGHT WAY: every handheld or used object must be
oriented exactly as its real-world use requires. A person reading,
watching or operating something (a phone, a photograph, a paper,
any device) has its functional side — screen, front, page — facing
THEIR OWN eyes; the camera then sees whatever side the staging
geometry implies (often its back). Show the functional side to the
camera ONLY when the shot text itself stages it toward the viewer.
Never flip, mirror or reverse an object's front and back.

CARRIED STATE (persist exactly — must match the neighbouring shots of this scene): The thick red circle now remains visible on the wall above the bed; Min-suk’s seated body, wounds, blood, and bloody footprints are still present in the room. Su-ri-young has taken the crumpled photograph from Min-suk’s hand.

NO PEOPLE IN THIS SHOT: the shot text shows only the place and its state — no living person or any body part appears in frame, unless the shot text itself explicitly says so.

No text, captions, watermarks or annotations anywhere.

Create ONE FINAL photorealistic live-action film still of the moment below — contemporary South Korea, 2026; all people are Korean unless stated. TIME OF DAY (lock): night.

SHOT TEXT (authoritative, Korean): 침대 위쪽 벽면에 붉은색의 굵은 원 형태가 선명하게 남아 있는 모습

LOCATION (lock): Inside the daughter's bedroom, on the wall directly above the bed in the dim nighttime apartment. The shot takes place here — the attached PREVIOUS SHOT STILL shows this exact place.

CAMERA & FRAME (follow exactly — this is the composition authority for this still; the reference images supply identity and place, never the framing):
- CAMERA: From low bedside height, the upward tilt passes the lower room and settles into a static, slightly low-angle view of the wall above the bed. The complete thick red circle sits in the upper-center at a restrained scale, with the bed edge retained along the lower frame to establish its physical placement rather than turning the mark into an abstract graphic.
- FRAMING SCALE: medium shot
- FRAME LAYOUT: wall with thick red circle above the bed in the upper-center of the frame, background; bed beneath the circle in the lower-center of the frame, midground.
- KEY BACKGROUND ELEMENTS: 침대 위 벽면 (Marked by a clearly visible red circle) — The camera-facing wall plane bears a complete thick red circle above the bed; used as Final supernatural focal surface; 침대 (Present beneath the circle) — Its upper edge is seen below the marked wall from the low bedside angle; used as Lower-frame spatial reference locating the mark.
Compose the frame exactly as specified above — camera angle, subject scale and screen placement. Keep true physical scale between people and background elements: fixtures and distant objects occupy only the small screen area their real size and distance dictate; never enlarge a background object into a foreground presence.

THIS SHOT CONTINUES THE PREVIOUS SHOT: everything the attached still established — the place, its fixed features and wear, each person's clothing and state — persists. Any person in it who cannot move stays PRECISELY as photographed (body, pose, contact points, held objects); only the camera changes.

REALIZE FIGURATIVE LANGUAGE AS A LIVE-ACTION SHOT: the Korean shot
text may describe characters metaphorically, figuratively or with
exaggeration. Photograph what a real movie camera would actually
record on a physical set — exaggerated or figurative impressions
become realistic staging choices (lighting, distance, angle,
wardrobe), not literal fantasy imagery.
EVERY CHARACTER IS A HUMAN BEING: unless the story explicitly
features non-human or virtual beings (as in science fiction or
fantasy), every character — however indirectly, vaguely or
figuratively the text describes them — IS a real human. When the
text gives no direct visual description of a person, IMAGINE one
and still show them as a concrete, fully-formed human being:
always render the human form to the maximum extent — build,
posture, face, hands, clothing — never reduce a person to a shape,
blob, solid silhouette or abstract mass.

EXPRESSIONS ARE ACTED, NEVER ANATOMICAL: when the text describes a
person's eyes, face or presence emotionally or figuratively (vacant,
hollow, dazed, burning, lifeless gaze and the like), it describes an
ACTOR'S PERFORMANCE captured by a real camera — realize it ONLY
through gaze direction, focus, eyelids, facial muscles, stillness
and posture. Every living person's eyes remain anatomically normal
human eyes with a natural iris and pupil, natural sclera and normal
proportions; NEVER whiten, blank out, cloud over, glow, enlarge or
otherwise alter eyeballs, skin or anatomy — unless the story
explicitly declares that being non-human or supernatural in form.

PROPS FACE THE RIGHT WAY: every handheld or used object must be
oriented exactly as its real-world use requires. A person reading,
watching or operating something (a phone, a photograph, a paper,
any device) has its functional side — screen, front, page — facing
THEIR OWN eyes; the camera then sees whatever side the staging
geometry implies (often its back). Show the functional side to the
camera ONLY when the shot text itself stages it toward the viewer.
Never flip, mirror or reverse an object's front and back.

CARRIED STATE (persist exactly — must match the neighbouring shots of this scene): The thick red circle now remains visible on the wall above the bed; Min-suk’s seated body, wounds, blood, and bloody footprints are still present in the room. Su-ri-young has taken the crumpled photograph from Min-suk’s hand.

NO PEOPLE IN THIS SHOT: the shot text shows only the place and its state — no living person or any body part appears in frame, unless the shot text itself explicitly says so.

No text, captions, watermarks or annotations anywhere.

VISUAL APPROACH — this take (camera, framing, composition):
Use an austere, nearly frontal composition with strong vertical and horizontal geometry, preserving the complete red circle at restrained scale in the upper center and the bed edge as a firm lower-frame anchor. Hold both wall texture and bed edge in clear focus under dim, soft nighttime illumination, with subdued surrounding color and controlled contrast so the red mark remains vivid without appearing luminous or graphic.

PRIORITY OF INSTRUCTIONS: the VISUAL APPROACH above governs camera,
framing and composition ONLY. Every other instruction in this prompt —
location, time of day, people and their identity, poses, props,
carried state, physical rules, and the no-text rule — is binding
exactly as written earlier. If the VISUAL APPROACH appears to conflict
with any of them, the earlier instruction wins.

Create ONE FINAL photorealistic live-action film still of the moment below — contemporary South Korea, 2026; all people are Korean unless stated. TIME OF DAY (lock): night.

SHOT TEXT (authoritative, Korean): 침대 위쪽 벽면에 붉은색의 굵은 원 형태가 선명하게 남아 있는 모습

LOCATION (lock): Inside the daughter's bedroom, on the wall directly above the bed in the dim nighttime apartment. The shot takes place here — the attached PREVIOUS SHOT STILL shows this exact place.

CAMERA & FRAME (follow exactly — this is the composition authority for this still; the reference images supply identity and place, never the framing):
- CAMERA: From low bedside height, the upward tilt passes the lower room and settles into a static, slightly low-angle view of the wall above the bed. The complete thick red circle sits in the upper-center at a restrained scale, with the bed edge retained along the lower frame to establish its physical placement rather than turning the mark into an abstract graphic.
- FRAMING SCALE: medium shot
- FRAME LAYOUT: wall with thick red circle above the bed in the upper-center of the frame, background; bed beneath the circle in the lower-center of the frame, midground.
- KEY BACKGROUND ELEMENTS: 침대 위 벽면 (Marked by a clearly visible red circle) — The camera-facing wall plane bears a complete thick red circle above the bed; used as Final supernatural focal surface; 침대 (Present beneath the circle) — Its upper edge is seen below the marked wall from the low bedside angle; used as Lower-frame spatial reference locating the mark.
Compose the frame exactly as specified above — camera angle, subject scale and screen placement. Keep true physical scale between people and background elements: fixtures and distant objects occupy only the small screen area their real size and distance dictate; never enlarge a background object into a foreground presence.

THIS SHOT CONTINUES THE PREVIOUS SHOT: everything the attached still established — the place, its fixed features and wear, each person's clothing and state — persists. Any person in it who cannot move stays PRECISELY as photographed (body, pose, contact points, held objects); only the camera changes.

REALIZE FIGURATIVE LANGUAGE AS A LIVE-ACTION SHOT: the Korean shot
text may describe characters metaphorically, figuratively or with
exaggeration. Photograph what a real movie camera would actually
record on a physical set — exaggerated or figurative impressions
become realistic staging choices (lighting, distance, angle,
wardrobe), not literal fantasy imagery.
EVERY CHARACTER IS A HUMAN BEING: unless the story explicitly
features non-human or virtual beings (as in science fiction or
fantasy), every character — however indirectly, vaguely or
figuratively the text describes them — IS a real human. When the
text gives no direct visual description of a person, IMAGINE one
and still show them as a concrete, fully-formed human being:
always render the human form to the maximum extent — build,
posture, face, hands, clothing — never reduce a person to a shape,
blob, solid silhouette or abstract mass.

EXPRESSIONS ARE ACTED, NEVER ANATOMICAL: when the text describes a
person's eyes, face or presence emotionally or figuratively (vacant,
hollow, dazed, burning, lifeless gaze and the like), it describes an
ACTOR'S PERFORMANCE captured by a real camera — realize it ONLY
through gaze direction, focus, eyelids, facial muscles, stillness
and posture. Every living person's eyes remain anatomically normal
human eyes with a natural iris and pupil, natural sclera and normal
proportions; NEVER whiten, blank out, cloud over, glow, enlarge or
otherwise alter eyeballs, skin or anatomy — unless the story
explicitly declares that being non-human or supernatural in form.

PROPS FACE THE RIGHT WAY: every handheld or used object must be
oriented exactly as its real-world use requires. A person reading,
watching or operating something (a phone, a photograph, a paper,
any device) has its functional side — screen, front, page — facing
THEIR OWN eyes; the camera then sees whatever side the staging
geometry implies (often its back). Show the functional side to the
camera ONLY when the shot text itself stages it toward the viewer.
Never flip, mirror or reverse an object's front and back.

CARRIED STATE (persist exactly — must match the neighbouring shots of this scene): The thick red circle now remains visible on the wall above the bed; Min-suk’s seated body, wounds, blood, and bloody footprints are still present in the room. Su-ri-young has taken the crumpled photograph from Min-suk’s hand.

NO PEOPLE IN THIS SHOT: the shot text shows only the place and its state — no living person or any body part appears in frame, unless the shot text itself explicitly says so.

No text, captions, watermarks or annotations anywhere.

VISUAL APPROACH — this take (camera, framing, composition):
Photograph from the locked low bedside angle with selective focus: render the wall and complete red circle sharply while allowing the lower bed edge to fall gently soft, still clearly readable as the spatial reference. Shape the dim nighttime light so it grazes the wall plane, revealing subtle surface texture and the physical thickness of the red mark; let the lower room recede into soft shadow for a tense, intimate sense of discovery.

PRIORITY OF INSTRUCTIONS: the VISUAL APPROACH above governs camera,
framing and composition ONLY. Every other instruction in this prompt —
location, time of day, people and their identity, poses, props,
carried state, physical rules, and the no-text rule — is binding
exactly as written earlier. If the VISUAL APPROACH appears to conflict
with any of them, the earlier instruction wins.

Create ONE FINAL photorealistic live-action film still of the moment below — contemporary South Korea, 2026; all people are Korean unless stated. TIME OF DAY (lock): night.

SHOT TEXT (authoritative, Korean): 혜수가 수리영의 어깨를 양손으로 쥔 채 마주 보고 선 정지 상태

LOCATION (lock): In the villa's nighttime courtyard at the base of the rooftop stairs, where the detective confronts the stunned daughter. The shot takes place here — the attached PREVIOUS SHOT STILL shows this exact place.

CAMERA & FRAME (follow exactly — this is the composition authority for this still; the reference images supply identity and place, never the framing):
- CAMERA: At chest height a few steps to the side, a static medium two-shot places 혜수 on the left and 수리영 on the right in opposing profiles, tightening enough to keep both of 혜수’s hands clearly visible on 수리영’s shoulders. Their faces and the locked hand-to-shoulder contact occupy the central band, while the staircase remains peripheral for the coming turn upward.
- FRAMING SCALE: medium shot
- KEY BACKGROUND ELEMENTS: 옥탑방 연결 계단 (Open for ascent) — The ascending side is visible beyond the pair near the edge of frame; used as Peripheral directional cue for 혜수’s subsequent movement.
Compose the frame exactly as specified above — camera angle, subject scale and screen placement. Keep true physical scale between people and background elements: fixtures and distant objects occupy only the small screen area their real size and distance dictate; never enlarge a background object into a foreground presence.

THIS SHOT CONTINUES THE PREVIOUS SHOT: everything the attached still established — the place, its fixed features and wear, each person's clothing and state — persists. Any person in it who cannot move stays PRECISELY as photographed (body, pose, contact points, held objects); only the camera changes.

PREVIOUS STILL USAGE (follow exactly — what to take from the attached still and what to exclude): Take the same nighttime courtyard, exterior stairs, rooftop entrance, and fixed architectural layout from the reference. Exclude the earlier solitary pose and any bicycle or approach action; include only the two women facing each other with hands on the younger woman’s shoulders.

REALIZE FIGURATIVE LANGUAGE AS A LIVE-ACTION SHOT: the Korean shot
text may describe characters metaphorically, figuratively or with
exaggeration. Photograph what a real movie camera would actually
record on a physical set — exaggerated or figurative impressions
become realistic staging choices (lighting, distance, angle,
wardrobe), not literal fantasy imagery.
EVERY CHARACTER IS A HUMAN BEING: unless the story explicitly
features non-human or virtual beings (as in science fiction or
fantasy), every character — however indirectly, vaguely or
figuratively the text describes them — IS a real human. When the
text gives no direct visual description of a person, IMAGINE one
and still show them as a concrete, fully-formed human being:
always render the human form to the maximum extent — build,
posture, face, hands, clothing — never reduce a person to a shape,
blob, solid silhouette or abstract mass.

EXPRESSIONS ARE ACTED, NEVER ANATOMICAL: when the text describes a
person's eyes, face or presence emotionally or figuratively (vacant,
hollow, dazed, burning, lifeless gaze and the like), it describes an
ACTOR'S PERFORMANCE captured by a real camera — realize it ONLY
through gaze direction, focus, eyelids, facial muscles, stillness
and posture. Every living person's eyes remain anatomically normal
human eyes with a natural iris and pupil, natural sclera and normal
proportions; NEVER whiten, blank out, cloud over, glow, enlarge or
otherwise alter eyeballs, skin or anatomy — unless the story
explicitly declares that being non-human or supernatural in form.

PROPS FACE THE RIGHT WAY: every handheld or used object must be
oriented exactly as its real-world use requires. A person reading,
watching or operating something (a phone, a photograph, a paper,
any device) has its functional side — screen, front, page — facing
THEIR OWN eyes; the camera then sees whatever side the staging
geometry implies (often its back). Show the functional side to the
camera ONLY when the shot text itself stages it toward the viewer.
Never flip, mirror or reverse an object's front and back.

PEOPLE: the SHOT TEXT alone decides whether any person is visible in this shot. IF a person appears, they must be one of: 혜수 (한국인 40대 초반 여성, 단정한 검은 머리, 성숙한 타원형 얼굴); 수리영 (한국인 20대 초반 여성, 긴 검은 생머리, 젊은 타원형 얼굴) — never anyone else, and never add a person the shot text does not show. IF only part of a person is in frame (a hand, arm, foot, back, silhouette), that body part belongs to the specific person the shot text names — its sex, age, build, skin and grooming must unmistakably match that person's profile above.

No text, captions, watermarks or annotations anywhere.

Create ONE FINAL photorealistic live-action film still of the moment below — contemporary South Korea, 2026; all people are Korean unless stated. TIME OF DAY (lock): night.

SHOT TEXT (authoritative, Korean): 혜수가 수리영의 어깨를 양손으로 쥔 채 마주 보고 선 정지 상태

LOCATION (lock): In the villa's nighttime courtyard at the base of the rooftop stairs, where the detective confronts the stunned daughter. The shot takes place here — the attached PREVIOUS SHOT STILL shows this exact place.

CAMERA & FRAME (follow exactly — this is the composition authority for this still; the reference images supply identity and place, never the framing):
- CAMERA: At chest height a few steps to the side, a static medium two-shot places 혜수 on the left and 수리영 on the right in opposing profiles, tightening enough to keep both of 혜수’s hands clearly visible on 수리영’s shoulders. Their faces and the locked hand-to-shoulder contact occupy the central band, while the staircase remains peripheral for the coming turn upward.
- FRAMING SCALE: medium shot
- KEY BACKGROUND ELEMENTS: 옥탑방 연결 계단 (Open for ascent) — The ascending side is visible beyond the pair near the edge of frame; used as Peripheral directional cue for 혜수’s subsequent movement.
Compose the frame exactly as specified above — camera angle, subject scale and screen placement. Keep true physical scale between people and background elements: fixtures and distant objects occupy only the small screen area their real size and distance dictate; never enlarge a background object into a foreground presence.

THIS SHOT CONTINUES THE PREVIOUS SHOT: everything the attached still established — the place, its fixed features and wear, each person's clothing and state — persists. Any person in it who cannot move stays PRECISELY as photographed (body, pose, contact points, held objects); only the camera changes.

PREVIOUS STILL USAGE (follow exactly — what to take from the attached still and what to exclude): Take the same nighttime courtyard, exterior stairs, rooftop entrance, and fixed architectural layout from the reference. Exclude the earlier solitary pose and any bicycle or approach action; include only the two women facing each other with hands on the younger woman’s shoulders.

REALIZE FIGURATIVE LANGUAGE AS A LIVE-ACTION SHOT: the Korean shot
text may describe characters metaphorically, figuratively or with
exaggeration. Photograph what a real movie camera would actually
record on a physical set — exaggerated or figurative impressions
become realistic staging choices (lighting, distance, angle,
wardrobe), not literal fantasy imagery.
EVERY CHARACTER IS A HUMAN BEING: unless the story explicitly
features non-human or virtual beings (as in science fiction or
fantasy), every character — however indirectly, vaguely or
figuratively the text describes them — IS a real human. When the
text gives no direct visual description of a person, IMAGINE one
and still show them as a concrete, fully-formed human being:
always render the human form to the maximum extent — build,
posture, face, hands, clothing — never reduce a person to a shape,
blob, solid silhouette or abstract mass.

EXPRESSIONS ARE ACTED, NEVER ANATOMICAL: when the text describes a
person's eyes, face or presence emotionally or figuratively (vacant,
hollow, dazed, burning, lifeless gaze and the like), it describes an
ACTOR'S PERFORMANCE captured by a real camera — realize it ONLY
through gaze direction, focus, eyelids, facial muscles, stillness
and posture. Every living person's eyes remain anatomically normal
human eyes with a natural iris and pupil, natural sclera and normal
proportions; NEVER whiten, blank out, cloud over, glow, enlarge or
otherwise alter eyeballs, skin or anatomy — unless the story
explicitly declares that being non-human or supernatural in form.

PROPS FACE THE RIGHT WAY: every handheld or used object must be
oriented exactly as its real-world use requires. A person reading,
watching or operating something (a phone, a photograph, a paper,
any device) has its functional side — screen, front, page — facing
THEIR OWN eyes; the camera then sees whatever side the staging
geometry implies (often its back). Show the functional side to the
camera ONLY when the shot text itself stages it toward the viewer.
Never flip, mirror or reverse an object's front and back.

PEOPLE: the SHOT TEXT alone decides whether any person is visible in this shot. IF a person appears, they must be one of: 혜수 (한국인 40대 초반 여성, 단정한 검은 머리, 성숙한 타원형 얼굴); 수리영 (한국인 20대 초반 여성, 긴 검은 생머리, 젊은 타원형 얼굴) — never anyone else, and never add a person the shot text does not show. IF only part of a person is in frame (a hand, arm, foot, back, silhouette), that body part belongs to the specific person the shot text names — its sex, age, build, skin and grooming must unmistakably match that person's profile above.

No text, captions, watermarks or annotations anywhere.

VISUAL APPROACH — this take (camera, framing, composition):
Use a gently compressed lens feel and shallow depth of field, holding both opposing profiles and the locked hand-to-shoulder contact in crisp focus while the courtyard architecture falls softly away. Shape the night light as restrained, directional side light that traces facial contours and separates dark hair from the background, with subdued contrast and dense shadows creating an intimate, suspended tension.

PRIORITY OF INSTRUCTIONS: the VISUAL APPROACH above governs camera,
framing and composition ONLY. Every other instruction in this prompt —
location, time of day, people and their identity, poses, props,
carried state, physical rules, and the no-text rule — is binding
exactly as written earlier. If the VISUAL APPROACH appears to conflict
with any of them, the earlier instruction wins.

Create ONE FINAL photorealistic live-action film still of the moment below — contemporary South Korea, 2026; all people are Korean unless stated. TIME OF DAY (lock): night.

SHOT TEXT (authoritative, Korean): 혜수가 수리영의 어깨를 양손으로 쥔 채 마주 보고 선 정지 상태

LOCATION (lock): In the villa's nighttime courtyard at the base of the rooftop stairs, where the detective confronts the stunned daughter. The shot takes place here — the attached PREVIOUS SHOT STILL shows this exact place.

CAMERA & FRAME (follow exactly — this is the composition authority for this still; the reference images supply identity and place, never the framing):
- CAMERA: At chest height a few steps to the side, a static medium two-shot places 혜수 on the left and 수리영 on the right in opposing profiles, tightening enough to keep both of 혜수’s hands clearly visible on 수리영’s shoulders. Their faces and the locked hand-to-shoulder contact occupy the central band, while the staircase remains peripheral for the coming turn upward.
- FRAMING SCALE: medium shot
- KEY BACKGROUND ELEMENTS: 옥탑방 연결 계단 (Open for ascent) — The ascending side is visible beyond the pair near the edge of frame; used as Peripheral directional cue for 혜수’s subsequent movement.
Compose the frame exactly as specified above — camera angle, subject scale and screen placement. Keep true physical scale between people and background elements: fixtures and distant objects occupy only the small screen area their real size and distance dictate; never enlarge a background object into a foreground presence.

THIS SHOT CONTINUES THE PREVIOUS SHOT: everything the attached still established — the place, its fixed features and wear, each person's clothing and state — persists. Any person in it who cannot move stays PRECISELY as photographed (body, pose, contact points, held objects); only the camera changes.

PREVIOUS STILL USAGE (follow exactly — what to take from the attached still and what to exclude): Take the same nighttime courtyard, exterior stairs, rooftop entrance, and fixed architectural layout from the reference. Exclude the earlier solitary pose and any bicycle or approach action; include only the two women facing each other with hands on the younger woman’s shoulders.

REALIZE FIGURATIVE LANGUAGE AS A LIVE-ACTION SHOT: the Korean shot
text may describe characters metaphorically, figuratively or with
exaggeration. Photograph what a real movie camera would actually
record on a physical set — exaggerated or figurative impressions
become realistic staging choices (lighting, distance, angle,
wardrobe), not literal fantasy imagery.
EVERY CHARACTER IS A HUMAN BEING: unless the story explicitly
features non-human or virtual beings (as in science fiction or
fantasy), every character — however indirectly, vaguely or
figuratively the text describes them — IS a real human. When the
text gives no direct visual description of a person, IMAGINE one
and still show them as a concrete, fully-formed human being:
always render the human form to the maximum extent — build,
posture, face, hands, clothing — never reduce a person to a shape,
blob, solid silhouette or abstract mass.

EXPRESSIONS ARE ACTED, NEVER ANATOMICAL: when the text describes a
person's eyes, face or presence emotionally or figuratively (vacant,
hollow, dazed, burning, lifeless gaze and the like), it describes an
ACTOR'S PERFORMANCE captured by a real camera — realize it ONLY
through gaze direction, focus, eyelids, facial muscles, stillness
and posture. Every living person's eyes remain anatomically normal
human eyes with a natural iris and pupil, natural sclera and normal
proportions; NEVER whiten, blank out, cloud over, glow, enlarge or
otherwise alter eyeballs, skin or anatomy — unless the story
explicitly declares that being non-human or supernatural in form.

PROPS FACE THE RIGHT WAY: every handheld or used object must be
oriented exactly as its real-world use requires. A person reading,
watching or operating something (a phone, a photograph, a paper,
any device) has its functional side — screen, front, page — facing
THEIR OWN eyes; the camera then sees whatever side the staging
geometry implies (often its back). Show the functional side to the
camera ONLY when the shot text itself stages it toward the viewer.
Never flip, mirror or reverse an object's front and back.

PEOPLE: the SHOT TEXT alone decides whether any person is visible in this shot. IF a person appears, they must be one of: 혜수 (한국인 40대 초반 여성, 단정한 검은 머리, 성숙한 타원형 얼굴); 수리영 (한국인 20대 초반 여성, 긴 검은 생머리, 젊은 타원형 얼굴) — never anyone else, and never add a person the shot text does not show. IF only part of a person is in frame (a hand, arm, foot, back, silhouette), that body part belongs to the specific person the shot text names — its sex, age, build, skin and grooming must unmistakably match that person's profile above.

No text, captions, watermarks or annotations anywhere.

VISUAL APPROACH — this take (camera, framing, composition):
Use a natural, slightly spacious lens feel with deeper focus, organizing the two profiles as a firm central barrier while preserving clear spatial depth into the courtyard and toward the peripheral staircase. Favor harder, motivated nighttime illumination with controlled pools of light and shadow across the architecture; keep the faces and both hands fully readable while the visible ascent at the frame edge exerts a quiet directional pull.

PRIORITY OF INSTRUCTIONS: the VISUAL APPROACH above governs camera,
framing and composition ONLY. Every other instruction in this prompt —
location, time of day, people and their identity, poses, props,
carried state, physical rules, and the no-text rule — is binding
exactly as written earlier. If the VISUAL APPROACH appears to conflict
with any of them, the earlier instruction wins.

Edit this photograph. Apply ONLY the corrections listed below,
each in place, changing nothing else:

CORRECTIONS:
- Remove the disembodied hand from Suriyoung's shoulder and reposition Hyesu's right hand to grasp the shoulder instead of the bicep.
- Adjust the pose to make Hyesu's left hand clearly visible resting on Suriyoung's far shoulder.
- Remove the watch from Hyesu's right wrist.

PRESERVE EVERYTHING ELSE EXACTLY as in the original photograph —
every person, pose, face, garment, object, set feature, lighting,
camera and framing stay identical. No text, captions or
watermarks anywhere.

Create ONE FINAL photorealistic live-action film still of the moment below — contemporary South Korea, 2026; all people are Korean unless stated. TIME OF DAY (lock): night.

SHOT TEXT (authoritative, Korean): 방 안 한가운데 서서 핏기 없는 얼굴로 멍하니 굳어 있는 수리영 클로즈업

LOCATION (lock): In the middle of the daughter's spotless bedroom, with no blood, wall mark, or body remaining. The shot takes place here — the attached PREVIOUS SHOT STILL shows this exact place.

CAMERA & FRAME (follow exactly — this is the composition authority for this still; the reference images supply identity and place, never the framing):
- CAMERA: Continue on the same eye-level room axis and complete the dolly-in on 수리영, changing only the camera distance from the preceding wide reveal; her pale face occupies the central portion while fragments of the orderly room remain at the edges. Hold her in a slight three-quarter angle with shallow focus, ending close enough that her locked expression becomes the sole point of attention before the camera arcs past her shoulder.
- FRAMING SCALE: close-up
- FRAME LAYOUT: 수리영 in the middle-center of the frame, foreground.
- KEY BACKGROUND ELEMENTS: clean bedroom floor (Clean and showing no bloodstains); used as Falls softly out of focus behind 수리영, preserving continuity with the preceding absence reveal; curtain area (No body remains behind the curtain) — Only part of its room-facing side remains visible behind 수리영; used as Remains as a peripheral background reminder of the vanished evidence.
Compose the frame exactly as specified above — camera angle, subject scale and screen placement. Keep true physical scale between people and background elements: fixtures and distant objects occupy only the small screen area their real size and distance dictate; never enlarge a background object into a foreground presence.

THIS SHOT CONTINUES THE PREVIOUS SHOT: everything the attached still established — the place, its fixed features and wear, each person's clothing and state — persists. Any person in it who cannot move stays PRECISELY as photographed (body, pose, contact points, held objects); only the camera changes.

PREVIOUS STILL USAGE (follow exactly — what to take from the attached still and what to exclude): Take the spotless bedroom layout, empty curtain area, floor, and unchanged nighttime lighting from the reference. Exclude the empty-room wide composition and do not add the missing corpse, blood, or attack debris; frame only the pale woman standing in the room.

REALIZE FIGURATIVE LANGUAGE AS A LIVE-ACTION SHOT: the Korean shot
text may describe characters metaphorically, figuratively or with
exaggeration. Photograph what a real movie camera would actually
record on a physical set — exaggerated or figurative impressions
become realistic staging choices (lighting, distance, angle,
wardrobe), not literal fantasy imagery.
EVERY CHARACTER IS A HUMAN BEING: unless the story explicitly
features non-human or virtual beings (as in science fiction or
fantasy), every character — however indirectly, vaguely or
figuratively the text describes them — IS a real human. When the
text gives no direct visual description of a person, IMAGINE one
and still show them as a concrete, fully-formed human being:
always render the human form to the maximum extent — build,
posture, face, hands, clothing — never reduce a person to a shape,
blob, solid silhouette or abstract mass.

EXPRESSIONS ARE ACTED, NEVER ANATOMICAL: when the text describes a
person's eyes, face or presence emotionally or figuratively (vacant,
hollow, dazed, burning, lifeless gaze and the like), it describes an
ACTOR'S PERFORMANCE captured by a real camera — realize it ONLY
through gaze direction, focus, eyelids, facial muscles, stillness
and posture. Every living person's eyes remain anatomically normal
human eyes with a natural iris and pupil, natural sclera and normal
proportions; NEVER whiten, blank out, cloud over, glow, enlarge or
otherwise alter eyeballs, skin or anatomy — unless the story
explicitly declares that being non-human or supernatural in form.

PROPS FACE THE RIGHT WAY: every handheld or used object must be
oriented exactly as its real-world use requires. A person reading,
watching or operating something (a phone, a photograph, a paper,
any device) has its functional side — screen, front, page — facing
THEIR OWN eyes; the camera then sees whatever side the staging
geometry implies (often its back). Show the functional side to the
camera ONLY when the shot text itself stages it toward the viewer.
Never flip, mirror or reverse an object's front and back.

CARRIED STATE (persist exactly — must match the neighbouring shots of this scene): The room remains excessively neat and completely free of the corpse, blood, and red circle Su-ri-young previously saw.

PEOPLE: the SHOT TEXT alone decides whether any person is visible in this shot. IF a person appears, they must be one of: 수리영 (한국인 20대 초반 여성, 긴 검은 생머리, 젊은 타원형 얼굴) — never anyone else, and never add a person the shot text does not show. IF only part of a person is in frame (a hand, arm, foot, back, silhouette), that body part belongs to the specific person the shot text names — its sex, age, build, skin and grooming must unmistakably match that person's profile above.

No text, captions, watermarks or annotations anywhere.

Create ONE FINAL photorealistic live-action film still of the moment below — contemporary South Korea, 2026; all people are Korean unless stated. TIME OF DAY (lock): night.

SHOT TEXT (authoritative, Korean): 방 안 한가운데 서서 핏기 없는 얼굴로 멍하니 굳어 있는 수리영 클로즈업

LOCATION (lock): In the middle of the daughter's spotless bedroom, with no blood, wall mark, or body remaining. The shot takes place here — the attached PREVIOUS SHOT STILL shows this exact place.

CAMERA & FRAME (follow exactly — this is the composition authority for this still; the reference images supply identity and place, never the framing):
- CAMERA: Continue on the same eye-level room axis and complete the dolly-in on 수리영, changing only the camera distance from the preceding wide reveal; her pale face occupies the central portion while fragments of the orderly room remain at the edges. Hold her in a slight three-quarter angle with shallow focus, ending close enough that her locked expression becomes the sole point of attention before the camera arcs past her shoulder.
- FRAMING SCALE: close-up
- FRAME LAYOUT: 수리영 in the middle-center of the frame, foreground.
- KEY BACKGROUND ELEMENTS: clean bedroom floor (Clean and showing no bloodstains); used as Falls softly out of focus behind 수리영, preserving continuity with the preceding absence reveal; curtain area (No body remains behind the curtain) — Only part of its room-facing side remains visible behind 수리영; used as Remains as a peripheral background reminder of the vanished evidence.
Compose the frame exactly as specified above — camera angle, subject scale and screen placement. Keep true physical scale between people and background elements: fixtures and distant objects occupy only the small screen area their real size and distance dictate; never enlarge a background object into a foreground presence.

THIS SHOT CONTINUES THE PREVIOUS SHOT: everything the attached still established — the place, its fixed features and wear, each person's clothing and state — persists. Any person in it who cannot move stays PRECISELY as photographed (body, pose, contact points, held objects); only the camera changes.

PREVIOUS STILL USAGE (follow exactly — what to take from the attached still and what to exclude): Take the spotless bedroom layout, empty curtain area, floor, and unchanged nighttime lighting from the reference. Exclude the empty-room wide composition and do not add the missing corpse, blood, or attack debris; frame only the pale woman standing in the room.

REALIZE FIGURATIVE LANGUAGE AS A LIVE-ACTION SHOT: the Korean shot
text may describe characters metaphorically, figuratively or with
exaggeration. Photograph what a real movie camera would actually
record on a physical set — exaggerated or figurative impressions
become realistic staging choices (lighting, distance, angle,
wardrobe), not literal fantasy imagery.
EVERY CHARACTER IS A HUMAN BEING: unless the story explicitly
features non-human or virtual beings (as in science fiction or
fantasy), every character — however indirectly, vaguely or
figuratively the text describes them — IS a real human. When the
text gives no direct visual description of a person, IMAGINE one
and still show them as a concrete, fully-formed human being:
always render the human form to the maximum extent — build,
posture, face, hands, clothing — never reduce a person to a shape,
blob, solid silhouette or abstract mass.

EXPRESSIONS ARE ACTED, NEVER ANATOMICAL: when the text describes a
person's eyes, face or presence emotionally or figuratively (vacant,
hollow, dazed, burning, lifeless gaze and the like), it describes an
ACTOR'S PERFORMANCE captured by a real camera — realize it ONLY
through gaze direction, focus, eyelids, facial muscles, stillness
and posture. Every living person's eyes remain anatomically normal
human eyes with a natural iris and pupil, natural sclera and normal
proportions; NEVER whiten, blank out, cloud over, glow, enlarge or
otherwise alter eyeballs, skin or anatomy — unless the story
explicitly declares that being non-human or supernatural in form.

PROPS FACE THE RIGHT WAY: every handheld or used object must be
oriented exactly as its real-world use requires. A person reading,
watching or operating something (a phone, a photograph, a paper,
any device) has its functional side — screen, front, page — facing
THEIR OWN eyes; the camera then sees whatever side the staging
geometry implies (often its back). Show the functional side to the
camera ONLY when the shot text itself stages it toward the viewer.
Never flip, mirror or reverse an object's front and back.

CARRIED STATE (persist exactly — must match the neighbouring shots of this scene): The room remains excessively neat and completely free of the corpse, blood, and red circle Su-ri-young previously saw.

PEOPLE: the SHOT TEXT alone decides whether any person is visible in this shot. IF a person appears, they must be one of: 수리영 (한국인 20대 초반 여성, 긴 검은 생머리, 젊은 타원형 얼굴) — never anyone else, and never add a person the shot text does not show. IF only part of a person is in frame (a hand, arm, foot, back, silhouette), that body part belongs to the specific person the shot text names — its sex, age, build, skin and grooming must unmistakably match that person's profile above.

No text, captions, watermarks or annotations anywhere.

VISUAL APPROACH — this take (camera, framing, composition):
Use a restrained, compressed lens feel at eye level, holding Su-ri-young rigidly in the middle-center with her face in a slight three-quarter angle. Frame the close-up with quiet, nearly symmetrical fragments of the orderly room at the edges, while the clean floor and curtain area dissolve into soft, shallow-focus planes behind her. Preserve the established nighttime illumination with gentle facial falloff, exposing for her pallor and minute tension in her unfocused gaze and still facial muscles; the image should feel clinically motionless and emotionally sealed.

PRIORITY OF INSTRUCTIONS: the VISUAL APPROACH above governs camera,
framing and composition ONLY. Every other instruction in this prompt —
location, time of day, people and their identity, poses, props,
carried state, physical rules, and the no-text rule — is binding
exactly as written earlier. If the VISUAL APPROACH appears to conflict
with any of them, the earlier instruction wins.

Create ONE FINAL photorealistic live-action film still of the moment below — contemporary South Korea, 2026; all people are Korean unless stated. TIME OF DAY (lock): night.

SHOT TEXT (authoritative, Korean): 방 안 한가운데 서서 핏기 없는 얼굴로 멍하니 굳어 있는 수리영 클로즈업

LOCATION (lock): In the middle of the daughter's spotless bedroom, with no blood, wall mark, or body remaining. The shot takes place here — the attached PREVIOUS SHOT STILL shows this exact place.

CAMERA & FRAME (follow exactly — this is the composition authority for this still; the reference images supply identity and place, never the framing):
- CAMERA: Continue on the same eye-level room axis and complete the dolly-in on 수리영, changing only the camera distance from the preceding wide reveal; her pale face occupies the central portion while fragments of the orderly room remain at the edges. Hold her in a slight three-quarter angle with shallow focus, ending close enough that her locked expression becomes the sole point of attention before the camera arcs past her shoulder.
- FRAMING SCALE: close-up
- FRAME LAYOUT: 수리영 in the middle-center of the frame, foreground.
- KEY BACKGROUND ELEMENTS: clean bedroom floor (Clean and showing no bloodstains); used as Falls softly out of focus behind 수리영, preserving continuity with the preceding absence reveal; curtain area (No body remains behind the curtain) — Only part of its room-facing side remains visible behind 수리영; used as Remains as a peripheral background reminder of the vanished evidence.
Compose the frame exactly as specified above — camera angle, subject scale and screen placement. Keep true physical scale between people and background elements: fixtures and distant objects occupy only the small screen area their real size and distance dictate; never enlarge a background object into a foreground presence.

THIS SHOT CONTINUES THE PREVIOUS SHOT: everything the attached still established — the place, its fixed features and wear, each person's clothing and state — persists. Any person in it who cannot move stays PRECISELY as photographed (body, pose, contact points, held objects); only the camera changes.

PREVIOUS STILL USAGE (follow exactly — what to take from the attached still and what to exclude): Take the spotless bedroom layout, empty curtain area, floor, and unchanged nighttime lighting from the reference. Exclude the empty-room wide composition and do not add the missing corpse, blood, or attack debris; frame only the pale woman standing in the room.

REALIZE FIGURATIVE LANGUAGE AS A LIVE-ACTION SHOT: the Korean shot
text may describe characters metaphorically, figuratively or with
exaggeration. Photograph what a real movie camera would actually
record on a physical set — exaggerated or figurative impressions
become realistic staging choices (lighting, distance, angle,
wardrobe), not literal fantasy imagery.
EVERY CHARACTER IS A HUMAN BEING: unless the story explicitly
features non-human or virtual beings (as in science fiction or
fantasy), every character — however indirectly, vaguely or
figuratively the text describes them — IS a real human. When the
text gives no direct visual description of a person, IMAGINE one
and still show them as a concrete, fully-formed human being:
always render the human form to the maximum extent — build,
posture, face, hands, clothing — never reduce a person to a shape,
blob, solid silhouette or abstract mass.

EXPRESSIONS ARE ACTED, NEVER ANATOMICAL: when the text describes a
person's eyes, face or presence emotionally or figuratively (vacant,
hollow, dazed, burning, lifeless gaze and the like), it describes an
ACTOR'S PERFORMANCE captured by a real camera — realize it ONLY
through gaze direction, focus, eyelids, facial muscles, stillness
and posture. Every living person's eyes remain anatomically normal
human eyes with a natural iris and pupil, natural sclera and normal
proportions; NEVER whiten, blank out, cloud over, glow, enlarge or
otherwise alter eyeballs, skin or anatomy — unless the story
explicitly declares that being non-human or supernatural in form.

PROPS FACE THE RIGHT WAY: every handheld or used object must be
oriented exactly as its real-world use requires. A person reading,
watching or operating something (a phone, a photograph, a paper,
any device) has its functional side — screen, front, page — facing
THEIR OWN eyes; the camera then sees whatever side the staging
geometry implies (often its back). Show the functional side to the
camera ONLY when the shot text itself stages it toward the viewer.
Never flip, mirror or reverse an object's front and back.

CARRIED STATE (persist exactly — must match the neighbouring shots of this scene): The room remains excessively neat and completely free of the corpse, blood, and red circle Su-ri-young previously saw.

PEOPLE: the SHOT TEXT alone decides whether any person is visible in this shot. IF a person appears, they must be one of: 수리영 (한국인 20대 초반 여성, 긴 검은 생머리, 젊은 타원형 얼굴) — never anyone else, and never add a person the shot text does not show. IF only part of a person is in frame (a hand, arm, foot, back, silhouette), that body part belongs to the specific person the shot text names — its sex, age, build, skin and grooming must unmistakably match that person's profile above.

No text, captions, watermarks or annotations anywhere.

VISUAL APPROACH — this take (camera, framing, composition):
Photograph the completed dolly-in with a more intimate, spatial lens feel, still eye level and close-up, placing Su-ri-young middle-center in the required slight three-quarter angle while allowing one near shoulder to give the frame a subtle directional pull toward the coming arc. Let the peripheral room fragments separate into layered depth: a soft trace of the spotless floor below and the empty curtain area receding behind her, both unmistakably present but visually subordinate. Keep the unchanged nighttime light natural to the room, with delicate tonal contrast across her pale face and extremely shallow focus pinning attention to her frozen eyes and barely engaged facial muscles.

PRIORITY OF INSTRUCTIONS: the VISUAL APPROACH above governs camera,
framing and composition ONLY. Every other instruction in this prompt —
location, time of day, people and their identity, poses, props,
carried state, physical rules, and the no-text rule — is binding
exactly as written earlier. If the VISUAL APPROACH appears to conflict
with any of them, the earlier instruction wins.

Edit this photograph. Apply ONLY the corrections listed below,
each in place, changing nothing else:

CORRECTIONS:
- Add the green cap to her head.
- Add the dark backpack straps running down over the front of her shoulders.
- Drape her long black hair down over the front of her shoulders.

PRESERVE EVERYTHING ELSE EXACTLY as in the original photograph —
every person, pose, face, garment, object, set feature, lighting,
camera and framing stay identical. No text, captions or
watermarks anywhere.

Create ONE FINAL photorealistic live-action film still of the moment below — contemporary South Korea, 2026; all people are Korean unless stated. TIME OF DAY (lock): night.

SHOT TEXT (authoritative, Korean): 혜수의 손가락이 휴대폰 화면의 전화번호를 김형사 눈앞에 내민 클로즈업

LOCATION (lock): Inside the nighttime police office, beside the duty desk where the detective shows a stored phone number to her colleague. The shot takes place here — the attached PREVIOUS SHOT STILL shows this exact place.

CAMERA & FRAME (follow exactly — this is the composition authority for this still; the reference images supply identity and place, never the framing):
- CAMERA: At 김형사's seated eye height, finish the dolly-in beside his eyeline so 혜수's fingertip and the displayed phone number occupy the central close focal area while his eye and cheek remain at one edge. 혜수's extended hand enters from the opposite side, keeping enough of both figures to establish who presents the phone and who examines it without turning the image into an isolated object insert.
- FRAMING SCALE: close-up
- KEY BACKGROUND ELEMENTS: mobile phone screen (On and displaying the stored number) — The lit display face is angled toward 김형사 and the camera, showing 강민숙's stored phone number; used as Primary evidence surface held between the two figures; 김형사's desk (In use during his duty shift) — Only the camera-facing edge and a small portion of the upper side are visible; used as Provides a partial lower-edge spatial reference for 김형사's seated position.
Compose the frame exactly as specified above — camera angle, subject scale and screen placement. Keep true physical scale between people and background elements: fixtures and distant objects occupy only the small screen area their real size and distance dictate; never enlarge a background object into a foreground presence.

THIS SHOT CONTINUES THE PREVIOUS SHOT: everything the attached still established — the place, its fixed features and wear, each person's clothing and state — persists. Any person in it who cannot move stays PRECISELY as photographed (body, pose, contact points, held objects); only the camera changes.

PREVIOUS STILL USAGE (follow exactly — what to take from the attached still and what to exclude): Take the established police-office materials, nighttime lighting, doorway context, and the officer’s continuing clothing. Exclude her doorway pose and face as compositional elements; isolate the hand and phone held before the male detective.

REALIZE FIGURATIVE LANGUAGE AS A LIVE-ACTION SHOT: the Korean shot
text may describe characters metaphorically, figuratively or with
exaggeration. Photograph what a real movie camera would actually
record on a physical set — exaggerated or figurative impressions
become realistic staging choices (lighting, distance, angle,
wardrobe), not literal fantasy imagery.
EVERY CHARACTER IS A HUMAN BEING: unless the story explicitly
features non-human or virtual beings (as in science fiction or
fantasy), every character — however indirectly, vaguely or
figuratively the text describes them — IS a real human. When the
text gives no direct visual description of a person, IMAGINE one
and still show them as a concrete, fully-formed human being:
always render the human form to the maximum extent — build,
posture, face, hands, clothing — never reduce a person to a shape,
blob, solid silhouette or abstract mass.

EXPRESSIONS ARE ACTED, NEVER ANATOMICAL: when the text describes a
person's eyes, face or presence emotionally or figuratively (vacant,
hollow, dazed, burning, lifeless gaze and the like), it describes an
ACTOR'S PERFORMANCE captured by a real camera — realize it ONLY
through gaze direction, focus, eyelids, facial muscles, stillness
and posture. Every living person's eyes remain anatomically normal
human eyes with a natural iris and pupil, natural sclera and normal
proportions; NEVER whiten, blank out, cloud over, glow, enlarge or
otherwise alter eyeballs, skin or anatomy — unless the story
explicitly declares that being non-human or supernatural in form.

PROPS FACE THE RIGHT WAY: every handheld or used object must be
oriented exactly as its real-world use requires. A person reading,
watching or operating something (a phone, a photograph, a paper,
any device) has its functional side — screen, front, page — facing
THEIR OWN eyes; the camera then sees whatever side the staging
geometry implies (often its back). Show the functional side to the
camera ONLY when the shot text itself stages it toward the viewer.
Never flip, mirror or reverse an object's front and back.

CARRIED STATE (persist exactly — must match the neighbouring shots of this scene): Hye-su continues holding the same phone, displaying Min-suk’s saved number to Detective Kim for tracing.

PEOPLE: the SHOT TEXT alone decides whether any person is visible in this shot. IF a person appears, they must be one of: 혜수 (한국인 40대 초반 여성, 단정한 검은 머리, 성숙한 타원형 얼굴); 김형사 (한국인 30대 남성, 짧은 검은 머리, 단정한 머리 모양, 성숙한 얼굴) — never anyone else, and never add a person the shot text does not show. IF only part of a person is in frame (a hand, arm, foot, back, silhouette), that body part belongs to the specific person the shot text names — its sex, age, build, skin and grooming must unmistakably match that person's profile above.

No text, captions, watermarks or annotations anywhere.

Create ONE FINAL photorealistic live-action film still of the moment below — contemporary South Korea, 2026; all people are Korean unless stated. TIME OF DAY (lock): night.

SHOT TEXT (authoritative, Korean): 혜수의 손가락이 휴대폰 화면의 전화번호를 김형사 눈앞에 내민 클로즈업

LOCATION (lock): Inside the nighttime police office, beside the duty desk where the detective shows a stored phone number to her colleague. The shot takes place here — the attached PREVIOUS SHOT STILL shows this exact place.

CAMERA & FRAME (follow exactly — this is the composition authority for this still; the reference images supply identity and place, never the framing):
- CAMERA: At 김형사's seated eye height, finish the dolly-in beside his eyeline so 혜수's fingertip and the displayed phone number occupy the central close focal area while his eye and cheek remain at one edge. 혜수's extended hand enters from the opposite side, keeping enough of both figures to establish who presents the phone and who examines it without turning the image into an isolated object insert.
- FRAMING SCALE: close-up
- KEY BACKGROUND ELEMENTS: mobile phone screen (On and displaying the stored number) — The lit display face is angled toward 김형사 and the camera, showing 강민숙's stored phone number; used as Primary evidence surface held between the two figures; 김형사's desk (In use during his duty shift) — Only the camera-facing edge and a small portion of the upper side are visible; used as Provides a partial lower-edge spatial reference for 김형사's seated position.
Compose the frame exactly as specified above — camera angle, subject scale and screen placement. Keep true physical scale between people and background elements: fixtures and distant objects occupy only the small screen area their real size and distance dictate; never enlarge a background object into a foreground presence.

THIS SHOT CONTINUES THE PREVIOUS SHOT: everything the attached still established — the place, its fixed features and wear, each person's clothing and state — persists. Any person in it who cannot move stays PRECISELY as photographed (body, pose, contact points, held objects); only the camera changes.

PREVIOUS STILL USAGE (follow exactly — what to take from the attached still and what to exclude): Take the established police-office materials, nighttime lighting, doorway context, and the officer’s continuing clothing. Exclude her doorway pose and face as compositional elements; isolate the hand and phone held before the male detective.

REALIZE FIGURATIVE LANGUAGE AS A LIVE-ACTION SHOT: the Korean shot
text may describe characters metaphorically, figuratively or with
exaggeration. Photograph what a real movie camera would actually
record on a physical set — exaggerated or figurative impressions
become realistic staging choices (lighting, distance, angle,
wardrobe), not literal fantasy imagery.
EVERY CHARACTER IS A HUMAN BEING: unless the story explicitly
features non-human or virtual beings (as in science fiction or
fantasy), every character — however indirectly, vaguely or
figuratively the text describes them — IS a real human. When the
text gives no direct visual description of a person, IMAGINE one
and still show them as a concrete, fully-formed human being:
always render the human form to the maximum extent — build,
posture, face, hands, clothing — never reduce a person to a shape,
blob, solid silhouette or abstract mass.

EXPRESSIONS ARE ACTED, NEVER ANATOMICAL: when the text describes a
person's eyes, face or presence emotionally or figuratively (vacant,
hollow, dazed, burning, lifeless gaze and the like), it describes an
ACTOR'S PERFORMANCE captured by a real camera — realize it ONLY
through gaze direction, focus, eyelids, facial muscles, stillness
and posture. Every living person's eyes remain anatomically normal
human eyes with a natural iris and pupil, natural sclera and normal
proportions; NEVER whiten, blank out, cloud over, glow, enlarge or
otherwise alter eyeballs, skin or anatomy — unless the story
explicitly declares that being non-human or supernatural in form.

PROPS FACE THE RIGHT WAY: every handheld or used object must be
oriented exactly as its real-world use requires. A person reading,
watching or operating something (a phone, a photograph, a paper,
any device) has its functional side — screen, front, page — facing
THEIR OWN eyes; the camera then sees whatever side the staging
geometry implies (often its back). Show the functional side to the
camera ONLY when the shot text itself stages it toward the viewer.
Never flip, mirror or reverse an object's front and back.

CARRIED STATE (persist exactly — must match the neighbouring shots of this scene): Hye-su continues holding the same phone, displaying Min-suk’s saved number to Detective Kim for tracing.

PEOPLE: the SHOT TEXT alone decides whether any person is visible in this shot. IF a person appears, they must be one of: 혜수 (한국인 40대 초반 여성, 단정한 검은 머리, 성숙한 타원형 얼굴); 김형사 (한국인 30대 남성, 짧은 검은 머리, 단정한 머리 모양, 성숙한 얼굴) — never anyone else, and never add a person the shot text does not show. IF only part of a person is in frame (a hand, arm, foot, back, silhouette), that body part belongs to the specific person the shot text names — its sex, age, build, skin and grooming must unmistakably match that person's profile above.

No text, captions, watermarks or annotations anywhere.

VISUAL APPROACH — this take (camera, framing, composition):
Use a tight, intimate lens feel with very shallow depth of field, resolving Hye-su’s fingertip and the illuminated stored number with crisp tactile detail while Detective Kim’s eye and cheek fall gently soft at the edge. Let the phone display provide a restrained cool glow against the established nighttime office light, making the evidence feel immediate and private. Keep the desk edge subdued and the background unobtrusive.

PRIORITY OF INSTRUCTIONS: the VISUAL APPROACH above governs camera,
framing and composition ONLY. Every other instruction in this prompt —
location, time of day, people and their identity, poses, props,
carried state, physical rules, and the no-text rule — is binding
exactly as written earlier. If the VISUAL APPROACH appears to conflict
with any of them, the earlier instruction wins.

Create ONE FINAL photorealistic live-action film still of the moment below — contemporary South Korea, 2026; all people are Korean unless stated. TIME OF DAY (lock): night.

SHOT TEXT (authoritative, Korean): 혜수의 손가락이 휴대폰 화면의 전화번호를 김형사 눈앞에 내민 클로즈업

LOCATION (lock): Inside the nighttime police office, beside the duty desk where the detective shows a stored phone number to her colleague. The shot takes place here — the attached PREVIOUS SHOT STILL shows this exact place.

CAMERA & FRAME (follow exactly — this is the composition authority for this still; the reference images supply identity and place, never the framing):
- CAMERA: At 김형사's seated eye height, finish the dolly-in beside his eyeline so 혜수's fingertip and the displayed phone number occupy the central close focal area while his eye and cheek remain at one edge. 혜수's extended hand enters from the opposite side, keeping enough of both figures to establish who presents the phone and who examines it without turning the image into an isolated object insert.
- FRAMING SCALE: close-up
- KEY BACKGROUND ELEMENTS: mobile phone screen (On and displaying the stored number) — The lit display face is angled toward 김형사 and the camera, showing 강민숙's stored phone number; used as Primary evidence surface held between the two figures; 김형사's desk (In use during his duty shift) — Only the camera-facing edge and a small portion of the upper side are visible; used as Provides a partial lower-edge spatial reference for 김형사's seated position.
Compose the frame exactly as specified above — camera angle, subject scale and screen placement. Keep true physical scale between people and background elements: fixtures and distant objects occupy only the small screen area their real size and distance dictate; never enlarge a background object into a foreground presence.

THIS SHOT CONTINUES THE PREVIOUS SHOT: everything the attached still established — the place, its fixed features and wear, each person's clothing and state — persists. Any person in it who cannot move stays PRECISELY as photographed (body, pose, contact points, held objects); only the camera changes.

PREVIOUS STILL USAGE (follow exactly — what to take from the attached still and what to exclude): Take the established police-office materials, nighttime lighting, doorway context, and the officer’s continuing clothing. Exclude her doorway pose and face as compositional elements; isolate the hand and phone held before the male detective.

REALIZE FIGURATIVE LANGUAGE AS A LIVE-ACTION SHOT: the Korean shot
text may describe characters metaphorically, figuratively or with
exaggeration. Photograph what a real movie camera would actually
record on a physical set — exaggerated or figurative impressions
become realistic staging choices (lighting, distance, angle,
wardrobe), not literal fantasy imagery.
EVERY CHARACTER IS A HUMAN BEING: unless the story explicitly
features non-human or virtual beings (as in science fiction or
fantasy), every character — however indirectly, vaguely or
figuratively the text describes them — IS a real human. When the
text gives no direct visual description of a person, IMAGINE one
and still show them as a concrete, fully-formed human being:
always render the human form to the maximum extent — build,
posture, face, hands, clothing — never reduce a person to a shape,
blob, solid silhouette or abstract mass.

EXPRESSIONS ARE ACTED, NEVER ANATOMICAL: when the text describes a
person's eyes, face or presence emotionally or figuratively (vacant,
hollow, dazed, burning, lifeless gaze and the like), it describes an
ACTOR'S PERFORMANCE captured by a real camera — realize it ONLY
through gaze direction, focus, eyelids, facial muscles, stillness
and posture. Every living person's eyes remain anatomically normal
human eyes with a natural iris and pupil, natural sclera and normal
proportions; NEVER whiten, blank out, cloud over, glow, enlarge or
otherwise alter eyeballs, skin or anatomy — unless the story
explicitly declares that being non-human or supernatural in form.

PROPS FACE THE RIGHT WAY: every handheld or used object must be
oriented exactly as its real-world use requires. A person reading,
watching or operating something (a phone, a photograph, a paper,
any device) has its functional side — screen, front, page — facing
THEIR OWN eyes; the camera then sees whatever side the staging
geometry implies (often its back). Show the functional side to the
camera ONLY when the shot text itself stages it toward the viewer.
Never flip, mirror or reverse an object's front and back.

CARRIED STATE (persist exactly — must match the neighbouring shots of this scene): Hye-su continues holding the same phone, displaying Min-suk’s saved number to Detective Kim for tracing.

PEOPLE: the SHOT TEXT alone decides whether any person is visible in this shot. IF a person appears, they must be one of: 혜수 (한국인 40대 초반 여성, 단정한 검은 머리, 성숙한 타원형 얼굴); 김형사 (한국인 30대 남성, 짧은 검은 머리, 단정한 머리 모양, 성숙한 얼굴) — never anyone else, and never add a person the shot text does not show. IF only part of a person is in frame (a hand, arm, foot, back, silhouette), that body part belongs to the specific person the shot text names — its sex, age, build, skin and grooming must unmistakably match that person's profile above.

No text, captions, watermarks or annotations anywhere.

VISUAL APPROACH — this take (camera, framing, composition):
Photograph the close-up with a compressed, observational lens feel and enough depth for both the displayed number and Detective Kim’s edge-framed eye to remain clearly readable. Arrange the phone, fingertip, and his fixed gaze as a taut visual line across the frame, emphasizing his act of verification rather than isolating the device. Preserve the established office illumination with controlled contrast and a faint falloff into the surrounding nighttime workspace.

PRIORITY OF INSTRUCTIONS: the VISUAL APPROACH above governs camera,
framing and composition ONLY. Every other instruction in this prompt —
location, time of day, people and their identity, poses, props,
carried state, physical rules, and the no-text rule — is binding
exactly as written earlier. If the VISUAL APPROACH appears to conflict
with any of them, the earlier instruction wins.

Edit this photograph. Apply ONLY the corrections listed below,
each in place, changing nothing else:

CORRECTIONS:
- Remove the disembodied fingers gripping the right edge of the phone and adjust the visible hand so it holds the device while pointing.

PRESERVE EVERYTHING ELSE EXACTLY as in the original photograph —
every person, pose, face, garment, object, set feature, lighting,
camera and framing stay identical. No text, captions or
watermarks anywhere.

Create ONE FINAL photorealistic live-action film still of the moment below — contemporary South Korea, 2026; all people are Korean unless stated. TIME OF DAY (lock): early morning.

SHOT TEXT (authoritative, Korean): 수리영이 휴대폰을 귀에 댄 채 굳은 표정으로 서 있는 상체

LOCATION (lock): Outside the supermarket near the exterior CCTV coverage area, where she calls the detective in daylight. The shot takes place here — the attached PREVIOUS SHOT STILL shows this exact place.

CAMERA & FRAME (follow exactly — this is the composition authority for this still; the reference images supply identity and place, never the framing):
- CAMERA: From outside the mart at Suriyoung's upper-body height, finish the prescribed arc behind her and settle into a stable frontal three-quarter medium framing, with her rigid face and phone occupying the center-right of frame. Keep the mart entrance softly present behind her and end all camera drift so the still composition bears the weight of her reaction.
- FRAMING SCALE: medium shot
- KEY BACKGROUND ELEMENTS: 마트 입구 (Visible from outside; open or closed state is not specified) — The exterior-facing side of the entrance is seen behind Suriyoung; used as Soft background context behind Suriyoung without competing with her expression.
Compose the frame exactly as specified above — camera angle, subject scale and screen placement. Keep true physical scale between people and background elements: fixtures and distant objects occupy only the small screen area their real size and distance dictate; never enlarge a background object into a foreground presence.

THIS SHOT CONTINUES THE PREVIOUS SHOT: everything the attached still established — the place, its fixed features and wear, each person's clothing and state — persists. Any person in it who cannot move stays PRECISELY as photographed (body, pose, contact points, held objects); only the camera changes.

PREVIOUS STILL USAGE (follow exactly — what to take from the attached still and what to exclude): Take the same market exterior wall, CCTV-camera placement, daylight, and the young woman’s continuing clothing. Exclude the male employee and the upward-looking two-person pose; show only the young woman making the phone call.

REALIZE FIGURATIVE LANGUAGE AS A LIVE-ACTION SHOT: the Korean shot
text may describe characters metaphorically, figuratively or with
exaggeration. Photograph what a real movie camera would actually
record on a physical set — exaggerated or figurative impressions
become realistic staging choices (lighting, distance, angle,
wardrobe), not literal fantasy imagery.
EVERY CHARACTER IS A HUMAN BEING: unless the story explicitly
features non-human or virtual beings (as in science fiction or
fantasy), every character — however indirectly, vaguely or
figuratively the text describes them — IS a real human. When the
text gives no direct visual description of a person, IMAGINE one
and still show them as a concrete, fully-formed human being:
always render the human form to the maximum extent — build,
posture, face, hands, clothing — never reduce a person to a shape,
blob, solid silhouette or abstract mass.

EXPRESSIONS ARE ACTED, NEVER ANATOMICAL: when the text describes a
person's eyes, face or presence emotionally or figuratively (vacant,
hollow, dazed, burning, lifeless gaze and the like), it describes an
ACTOR'S PERFORMANCE captured by a real camera — realize it ONLY
through gaze direction, focus, eyelids, facial muscles, stillness
and posture. Every living person's eyes remain anatomically normal
human eyes with a natural iris and pupil, natural sclera and normal
proportions; NEVER whiten, blank out, cloud over, glow, enlarge or
otherwise alter eyeballs, skin or anatomy — unless the story
explicitly declares that being non-human or supernatural in form.

PROPS FACE THE RIGHT WAY: every handheld or used object must be
oriented exactly as its real-world use requires. A person reading,
watching or operating something (a phone, a photograph, a paper,
any device) has its functional side — screen, front, page — facing
THEIR OWN eyes; the camera then sees whatever side the staging
geometry implies (often its back). Show the functional side to the
camera ONLY when the shot text itself stages it toward the viewer.
Never flip, mirror or reverse an object's front and back.

CARRIED STATE (persist exactly — must match the neighbouring shots of this scene): Su-ri-young still carries her doll-adorned backpack and holds her phone while explaining the CCTV location to Hye-su; her bicycle remains parked nearby.

PEOPLE: the SHOT TEXT alone decides whether any person is visible in this shot. IF a person appears, they must be one of: 수리영 (한국인 20대 초반 여성, 긴 검은 생머리, 젊은 타원형 얼굴) — never anyone else, and never add a person the shot text does not show. IF only part of a person is in frame (a hand, arm, foot, back, silhouette), that body part belongs to the specific person the shot text names — its sex, age, build, skin and grooming must unmistakably match that person's profile above.

No text, captions, watermarks or annotations anywhere.

Create ONE FINAL photorealistic live-action film still of the moment below — contemporary South Korea, 2026; all people are Korean unless stated. TIME OF DAY (lock): early morning.

SHOT TEXT (authoritative, Korean): 수리영이 휴대폰을 귀에 댄 채 굳은 표정으로 서 있는 상체

LOCATION (lock): Outside the supermarket near the exterior CCTV coverage area, where she calls the detective in daylight. The shot takes place here — the attached PREVIOUS SHOT STILL shows this exact place.

CAMERA & FRAME (follow exactly — this is the composition authority for this still; the reference images supply identity and place, never the framing):
- CAMERA: From outside the mart at Suriyoung's upper-body height, finish the prescribed arc behind her and settle into a stable frontal three-quarter medium framing, with her rigid face and phone occupying the center-right of frame. Keep the mart entrance softly present behind her and end all camera drift so the still composition bears the weight of her reaction.
- FRAMING SCALE: medium shot
- KEY BACKGROUND ELEMENTS: 마트 입구 (Visible from outside; open or closed state is not specified) — The exterior-facing side of the entrance is seen behind Suriyoung; used as Soft background context behind Suriyoung without competing with her expression.
Compose the frame exactly as specified above — camera angle, subject scale and screen placement. Keep true physical scale between people and background elements: fixtures and distant objects occupy only the small screen area their real size and distance dictate; never enlarge a background object into a foreground presence.

THIS SHOT CONTINUES THE PREVIOUS SHOT: everything the attached still established — the place, its fixed features and wear, each person's clothing and state — persists. Any person in it who cannot move stays PRECISELY as photographed (body, pose, contact points, held objects); only the camera changes.

PREVIOUS STILL USAGE (follow exactly — what to take from the attached still and what to exclude): Take the same market exterior wall, CCTV-camera placement, daylight, and the young woman’s continuing clothing. Exclude the male employee and the upward-looking two-person pose; show only the young woman making the phone call.

REALIZE FIGURATIVE LANGUAGE AS A LIVE-ACTION SHOT: the Korean shot
text may describe characters metaphorically, figuratively or with
exaggeration. Photograph what a real movie camera would actually
record on a physical set — exaggerated or figurative impressions
become realistic staging choices (lighting, distance, angle,
wardrobe), not literal fantasy imagery.
EVERY CHARACTER IS A HUMAN BEING: unless the story explicitly
features non-human or virtual beings (as in science fiction or
fantasy), every character — however indirectly, vaguely or
figuratively the text describes them — IS a real human. When the
text gives no direct visual description of a person, IMAGINE one
and still show them as a concrete, fully-formed human being:
always render the human form to the maximum extent — build,
posture, face, hands, clothing — never reduce a person to a shape,
blob, solid silhouette or abstract mass.

EXPRESSIONS ARE ACTED, NEVER ANATOMICAL: when the text describes a
person's eyes, face or presence emotionally or figuratively (vacant,
hollow, dazed, burning, lifeless gaze and the like), it describes an
ACTOR'S PERFORMANCE captured by a real camera — realize it ONLY
through gaze direction, focus, eyelids, facial muscles, stillness
and posture. Every living person's eyes remain anatomically normal
human eyes with a natural iris and pupil, natural sclera and normal
proportions; NEVER whiten, blank out, cloud over, glow, enlarge or
otherwise alter eyeballs, skin or anatomy — unless the story
explicitly declares that being non-human or supernatural in form.

PROPS FACE THE RIGHT WAY: every handheld or used object must be
oriented exactly as its real-world use requires. A person reading,
watching or operating something (a phone, a photograph, a paper,
any device) has its functional side — screen, front, page — facing
THEIR OWN eyes; the camera then sees whatever side the staging
geometry implies (often its back). Show the functional side to the
camera ONLY when the shot text itself stages it toward the viewer.
Never flip, mirror or reverse an object's front and back.

CARRIED STATE (persist exactly — must match the neighbouring shots of this scene): Su-ri-young still carries her doll-adorned backpack and holds her phone while explaining the CCTV location to Hye-su; her bicycle remains parked nearby.

PEOPLE: the SHOT TEXT alone decides whether any person is visible in this shot. IF a person appears, they must be one of: 수리영 (한국인 20대 초반 여성, 긴 검은 생머리, 젊은 타원형 얼굴) — never anyone else, and never add a person the shot text does not show. IF only part of a person is in frame (a hand, arm, foot, back, silhouette), that body part belongs to the specific person the shot text names — its sex, age, build, skin and grooming must unmistakably match that person's profile above.

No text, captions, watermarks or annotations anywhere.

VISUAL APPROACH — this take (camera, framing, composition):
Use a restrained long-lens feel with shallow depth of field, gently compressing the entrance into a soft, unobtrusive backdrop. Hold crisp focus on her rigid facial muscles, fixed gaze, and the phone at her ear, with subdued early-morning daylight grazing her face from the exterior side. Keep the image still and tightly controlled, emphasizing the suspended tension of her reaction.

PRIORITY OF INSTRUCTIONS: the VISUAL APPROACH above governs camera,
framing and composition ONLY. Every other instruction in this prompt —
location, time of day, people and their identity, poses, props,
carried state, physical rules, and the no-text rule — is binding
exactly as written earlier. If the VISUAL APPROACH appears to conflict
with any of them, the earlier instruction wins.

Create ONE FINAL photorealistic live-action film still of the moment below — contemporary South Korea, 2026; all people are Korean unless stated. TIME OF DAY (lock): early morning.

SHOT TEXT (authoritative, Korean): 수리영이 휴대폰을 귀에 댄 채 굳은 표정으로 서 있는 상체

LOCATION (lock): Outside the supermarket near the exterior CCTV coverage area, where she calls the detective in daylight. The shot takes place here — the attached PREVIOUS SHOT STILL shows this exact place.

CAMERA & FRAME (follow exactly — this is the composition authority for this still; the reference images supply identity and place, never the framing):
- CAMERA: From outside the mart at Suriyoung's upper-body height, finish the prescribed arc behind her and settle into a stable frontal three-quarter medium framing, with her rigid face and phone occupying the center-right of frame. Keep the mart entrance softly present behind her and end all camera drift so the still composition bears the weight of her reaction.
- FRAMING SCALE: medium shot
- KEY BACKGROUND ELEMENTS: 마트 입구 (Visible from outside; open or closed state is not specified) — The exterior-facing side of the entrance is seen behind Suriyoung; used as Soft background context behind Suriyoung without competing with her expression.
Compose the frame exactly as specified above — camera angle, subject scale and screen placement. Keep true physical scale between people and background elements: fixtures and distant objects occupy only the small screen area their real size and distance dictate; never enlarge a background object into a foreground presence.

THIS SHOT CONTINUES THE PREVIOUS SHOT: everything the attached still established — the place, its fixed features and wear, each person's clothing and state — persists. Any person in it who cannot move stays PRECISELY as photographed (body, pose, contact points, held objects); only the camera changes.

PREVIOUS STILL USAGE (follow exactly — what to take from the attached still and what to exclude): Take the same market exterior wall, CCTV-camera placement, daylight, and the young woman’s continuing clothing. Exclude the male employee and the upward-looking two-person pose; show only the young woman making the phone call.

REALIZE FIGURATIVE LANGUAGE AS A LIVE-ACTION SHOT: the Korean shot
text may describe characters metaphorically, figuratively or with
exaggeration. Photograph what a real movie camera would actually
record on a physical set — exaggerated or figurative impressions
become realistic staging choices (lighting, distance, angle,
wardrobe), not literal fantasy imagery.
EVERY CHARACTER IS A HUMAN BEING: unless the story explicitly
features non-human or virtual beings (as in science fiction or
fantasy), every character — however indirectly, vaguely or
figuratively the text describes them — IS a real human. When the
text gives no direct visual description of a person, IMAGINE one
and still show them as a concrete, fully-formed human being:
always render the human form to the maximum extent — build,
posture, face, hands, clothing — never reduce a person to a shape,
blob, solid silhouette or abstract mass.

EXPRESSIONS ARE ACTED, NEVER ANATOMICAL: when the text describes a
person's eyes, face or presence emotionally or figuratively (vacant,
hollow, dazed, burning, lifeless gaze and the like), it describes an
ACTOR'S PERFORMANCE captured by a real camera — realize it ONLY
through gaze direction, focus, eyelids, facial muscles, stillness
and posture. Every living person's eyes remain anatomically normal
human eyes with a natural iris and pupil, natural sclera and normal
proportions; NEVER whiten, blank out, cloud over, glow, enlarge or
otherwise alter eyeballs, skin or anatomy — unless the story
explicitly declares that being non-human or supernatural in form.

PROPS FACE THE RIGHT WAY: every handheld or used object must be
oriented exactly as its real-world use requires. A person reading,
watching or operating something (a phone, a photograph, a paper,
any device) has its functional side — screen, front, page — facing
THEIR OWN eyes; the camera then sees whatever side the staging
geometry implies (often its back). Show the functional side to the
camera ONLY when the shot text itself stages it toward the viewer.
Never flip, mirror or reverse an object's front and back.

CARRIED STATE (persist exactly — must match the neighbouring shots of this scene): Su-ri-young still carries her doll-adorned backpack and holds her phone while explaining the CCTV location to Hye-su; her bicycle remains parked nearby.

PEOPLE: the SHOT TEXT alone decides whether any person is visible in this shot. IF a person appears, they must be one of: 수리영 (한국인 20대 초반 여성, 긴 검은 생머리, 젊은 타원형 얼굴) — never anyone else, and never add a person the shot text does not show. IF only part of a person is in frame (a hand, arm, foot, back, silhouette), that body part belongs to the specific person the shot text names — its sex, age, build, skin and grooming must unmistakably match that person's profile above.

No text, captions, watermarks or annotations anywhere.

VISUAL APPROACH — this take (camera, framing, composition):
Use a natural, moderately wide perspective while preserving the prescribed medium scale, allowing more spatial separation between her and the softly legible mart entrance. Build the composition from the entrance geometry and restrained negative space, with her center-right placement feeling pinned against the environment. Favor even early-morning daylight, modest depth of field, and quiet tonal contrast so her absolute stillness carries the dramatic weight.

PRIORITY OF INSTRUCTIONS: the VISUAL APPROACH above governs camera,
framing and composition ONLY. Every other instruction in this prompt —
location, time of day, people and their identity, poses, props,
carried state, physical rules, and the no-text rule — is binding
exactly as written earlier. If the VISUAL APPROACH appears to conflict
with any of them, the earlier instruction wins.

Edit this photograph. Apply ONLY the corrections listed below,
each in place, changing nothing else:

CORRECTIONS:
- Replace the brick wall and electrical boxes directly behind the character with the mart entrance door and overhang from the reference.
- Remove the logo from the front of the green cap to make it completely plain.
- Change the phone's color and finish to match the dark grey/black appearance of the provided prop reference.

PRESERVE EVERYTHING ELSE EXACTLY as in the original photograph —
every person, pose, face, garment, object, set feature, lighting,
camera and framing stay identical. No text, captions or
watermarks anywhere.

Create ONE FINAL photorealistic live-action film still of the moment below — contemporary South Korea, 2026; all people are Korean unless stated. TIME OF DAY (lock): early morning.

SHOT TEXT (authoritative, Korean): 수리영이 두 손을 꽉 쥔 채 모니터를 뚫어지게 응시하는 굳은 얼굴 클로즈업

LOCATION (lock): Behind the seated staff at the CCTV desk inside the supermarket security office, under office lighting. The shot takes place here — the attached PREVIOUS SHOT STILL shows this exact place.

CAMERA & FRAME (follow exactly — this is the composition authority for this still; the reference images supply identity and place, never the framing):
- CAMERA: At Suriyoung's eye level on the monitor-facing side, complete the prescribed dolly inward until her rigid face occupies most of the close frame. Preserve her clasped hands along the lower edge and keep her gaze on the same axis, making camera distance the sole emphasized change from the wider setup.
- FRAMING SCALE: close-up
- FRAME LAYOUT: 수리영 in the middle-center of the frame, foreground, looks toward CCTV monitor.
- KEY BACKGROUND ELEMENTS: CCTV 모니터 (On and replaying footage) — The active front face is turned toward Suriyoung; its indistinct dark playback is only marginally visible from this angle; used as Held as a soft off-axis context for Suriyoung's fixed eyeline.
Compose the frame exactly as specified above — camera angle, subject scale and screen placement. Keep true physical scale between people and background elements: fixtures and distant objects occupy only the small screen area their real size and distance dictate; never enlarge a background object into a foreground presence.

THIS SHOT CONTINUES THE PREVIOUS SHOT: everything the attached still established — the place, its fixed features and wear, each person's clothing and state — persists. Any person in it who cannot move stays PRECISELY as photographed (body, pose, contact points, held objects); only the camera changes.

PREVIOUS STILL USAGE (follow exactly — what to take from the attached still and what to exclude): Take the same security-office lighting, monitor glow, desk arrangement, and the young woman’s position behind the seated staff. Exclude the seated security employee and officer from the close framing; focus on the young woman’s tense face and tightly clasped hands.

REALIZE FIGURATIVE LANGUAGE AS A LIVE-ACTION SHOT: the Korean shot
text may describe characters metaphorically, figuratively or with
exaggeration. Photograph what a real movie camera would actually
record on a physical set — exaggerated or figurative impressions
become realistic staging choices (lighting, distance, angle,
wardrobe), not literal fantasy imagery.
EVERY CHARACTER IS A HUMAN BEING: unless the story explicitly
features non-human or virtual beings (as in science fiction or
fantasy), every character — however indirectly, vaguely or
figuratively the text describes them — IS a real human. When the
text gives no direct visual description of a person, IMAGINE one
and still show them as a concrete, fully-formed human being:
always render the human form to the maximum extent — build,
posture, face, hands, clothing — never reduce a person to a shape,
blob, solid silhouette or abstract mass.

EXPRESSIONS ARE ACTED, NEVER ANATOMICAL: when the text describes a
person's eyes, face or presence emotionally or figuratively (vacant,
hollow, dazed, burning, lifeless gaze and the like), it describes an
ACTOR'S PERFORMANCE captured by a real camera — realize it ONLY
through gaze direction, focus, eyelids, facial muscles, stillness
and posture. Every living person's eyes remain anatomically normal
human eyes with a natural iris and pupil, natural sclera and normal
proportions; NEVER whiten, blank out, cloud over, glow, enlarge or
otherwise alter eyeballs, skin or anatomy — unless the story
explicitly declares that being non-human or supernatural in form.

PROPS FACE THE RIGHT WAY: every handheld or used object must be
oriented exactly as its real-world use requires. A person reading,
watching or operating something (a phone, a photograph, a paper,
any device) has its functional side — screen, front, page — facing
THEIR OWN eyes; the camera then sees whatever side the staging
geometry implies (often its back). Show the functional side to the
camera ONLY when the shot text itself stages it toward the viewer.
Never flip, mirror or reverse an object's front and back.

CARRIED STATE (persist exactly — must match the neighbouring shots of this scene): The same section of CCTV footage remains abnormally dark and unidentifiable on every replay. Su-ri-young’s doll-adorned backpack remains with her.

PEOPLE: the SHOT TEXT alone decides whether any person is visible in this shot. IF a person appears, they must be one of: 수리영 (한국인 20대 초반 여성, 긴 검은 생머리, 젊은 타원형 얼굴) — never anyone else, and never add a person the shot text does not show. IF only part of a person is in frame (a hand, arm, foot, back, silhouette), that body part belongs to the specific person the shot text names — its sex, age, build, skin and grooming must unmistakably match that person's profile above.

No text, captions, watermarks or annotations anywhere.

Create ONE FINAL photorealistic live-action film still of the moment below — contemporary South Korea, 2026; all people are Korean unless stated. TIME OF DAY (lock): early morning.

SHOT TEXT (authoritative, Korean): 수리영이 두 손을 꽉 쥔 채 모니터를 뚫어지게 응시하는 굳은 얼굴 클로즈업

LOCATION (lock): Behind the seated staff at the CCTV desk inside the supermarket security office, under office lighting. The shot takes place here — the attached PREVIOUS SHOT STILL shows this exact place.

CAMERA & FRAME (follow exactly — this is the composition authority for this still; the reference images supply identity and place, never the framing):
- CAMERA: At Suriyoung's eye level on the monitor-facing side, complete the prescribed dolly inward until her rigid face occupies most of the close frame. Preserve her clasped hands along the lower edge and keep her gaze on the same axis, making camera distance the sole emphasized change from the wider setup.
- FRAMING SCALE: close-up
- FRAME LAYOUT: 수리영 in the middle-center of the frame, foreground, looks toward CCTV monitor.
- KEY BACKGROUND ELEMENTS: CCTV 모니터 (On and replaying footage) — The active front face is turned toward Suriyoung; its indistinct dark playback is only marginally visible from this angle; used as Held as a soft off-axis context for Suriyoung's fixed eyeline.
Compose the frame exactly as specified above — camera angle, subject scale and screen placement. Keep true physical scale between people and background elements: fixtures and distant objects occupy only the small screen area their real size and distance dictate; never enlarge a background object into a foreground presence.

THIS SHOT CONTINUES THE PREVIOUS SHOT: everything the attached still established — the place, its fixed features and wear, each person's clothing and state — persists. Any person in it who cannot move stays PRECISELY as photographed (body, pose, contact points, held objects); only the camera changes.

PREVIOUS STILL USAGE (follow exactly — what to take from the attached still and what to exclude): Take the same security-office lighting, monitor glow, desk arrangement, and the young woman’s position behind the seated staff. Exclude the seated security employee and officer from the close framing; focus on the young woman’s tense face and tightly clasped hands.

REALIZE FIGURATIVE LANGUAGE AS A LIVE-ACTION SHOT: the Korean shot
text may describe characters metaphorically, figuratively or with
exaggeration. Photograph what a real movie camera would actually
record on a physical set — exaggerated or figurative impressions
become realistic staging choices (lighting, distance, angle,
wardrobe), not literal fantasy imagery.
EVERY CHARACTER IS A HUMAN BEING: unless the story explicitly
features non-human or virtual beings (as in science fiction or
fantasy), every character — however indirectly, vaguely or
figuratively the text describes them — IS a real human. When the
text gives no direct visual description of a person, IMAGINE one
and still show them as a concrete, fully-formed human being:
always render the human form to the maximum extent — build,
posture, face, hands, clothing — never reduce a person to a shape,
blob, solid silhouette or abstract mass.

EXPRESSIONS ARE ACTED, NEVER ANATOMICAL: when the text describes a
person's eyes, face or presence emotionally or figuratively (vacant,
hollow, dazed, burning, lifeless gaze and the like), it describes an
ACTOR'S PERFORMANCE captured by a real camera — realize it ONLY
through gaze direction, focus, eyelids, facial muscles, stillness
and posture. Every living person's eyes remain anatomically normal
human eyes with a natural iris and pupil, natural sclera and normal
proportions; NEVER whiten, blank out, cloud over, glow, enlarge or
otherwise alter eyeballs, skin or anatomy — unless the story
explicitly declares that being non-human or supernatural in form.

PROPS FACE THE RIGHT WAY: every handheld or used object must be
oriented exactly as its real-world use requires. A person reading,
watching or operating something (a phone, a photograph, a paper,
any device) has its functional side — screen, front, page — facing
THEIR OWN eyes; the camera then sees whatever side the staging
geometry implies (often its back). Show the functional side to the
camera ONLY when the shot text itself stages it toward the viewer.
Never flip, mirror or reverse an object's front and back.

CARRIED STATE (persist exactly — must match the neighbouring shots of this scene): The same section of CCTV footage remains abnormally dark and unidentifiable on every replay. Su-ri-young’s doll-adorned backpack remains with her.

PEOPLE: the SHOT TEXT alone decides whether any person is visible in this shot. IF a person appears, they must be one of: 수리영 (한국인 20대 초반 여성, 긴 검은 생머리, 젊은 타원형 얼굴) — never anyone else, and never add a person the shot text does not show. IF only part of a person is in frame (a hand, arm, foot, back, silhouette), that body part belongs to the specific person the shot text names — its sex, age, build, skin and grooming must unmistakably match that person's profile above.

No text, captions, watermarks or annotations anywhere.

VISUAL APPROACH — this take (camera, framing, composition):
Use an intimate, nearly frontal close-up with very shallow depth of field, holding Suriyoung’s rigid face in crisp focus while the marginal monitor presence falls into a soft, dark blur off axis. Let the office lighting remain restrained and natural, with the monitor glow subtly lifting her eyes and cheek planes; preserve her tightly clasped hands as a tense, slightly softer anchor along the lower edge. Favor stillness, symmetry, and minute facial tension for a suffocating sense of concentration.

PRIORITY OF INSTRUCTIONS: the VISUAL APPROACH above governs camera,
framing and composition ONLY. Every other instruction in this prompt —
location, time of day, people and their identity, poses, props,
carried state, physical rules, and the no-text rule — is binding
exactly as written earlier. If the VISUAL APPROACH appears to conflict
with any of them, the earlier instruction wins.

Create ONE FINAL photorealistic live-action film still of the moment below — contemporary South Korea, 2026; all people are Korean unless stated. TIME OF DAY (lock): early morning.

SHOT TEXT (authoritative, Korean): 수리영이 두 손을 꽉 쥔 채 모니터를 뚫어지게 응시하는 굳은 얼굴 클로즈업

LOCATION (lock): Behind the seated staff at the CCTV desk inside the supermarket security office, under office lighting. The shot takes place here — the attached PREVIOUS SHOT STILL shows this exact place.

CAMERA & FRAME (follow exactly — this is the composition authority for this still; the reference images supply identity and place, never the framing):
- CAMERA: At Suriyoung's eye level on the monitor-facing side, complete the prescribed dolly inward until her rigid face occupies most of the close frame. Preserve her clasped hands along the lower edge and keep her gaze on the same axis, making camera distance the sole emphasized change from the wider setup.
- FRAMING SCALE: close-up
- FRAME LAYOUT: 수리영 in the middle-center of the frame, foreground, looks toward CCTV monitor.
- KEY BACKGROUND ELEMENTS: CCTV 모니터 (On and replaying footage) — The active front face is turned toward Suriyoung; its indistinct dark playback is only marginally visible from this angle; used as Held as a soft off-axis context for Suriyoung's fixed eyeline.
Compose the frame exactly as specified above — camera angle, subject scale and screen placement. Keep true physical scale between people and background elements: fixtures and distant objects occupy only the small screen area their real size and distance dictate; never enlarge a background object into a foreground presence.

THIS SHOT CONTINUES THE PREVIOUS SHOT: everything the attached still established — the place, its fixed features and wear, each person's clothing and state — persists. Any person in it who cannot move stays PRECISELY as photographed (body, pose, contact points, held objects); only the camera changes.

PREVIOUS STILL USAGE (follow exactly — what to take from the attached still and what to exclude): Take the same security-office lighting, monitor glow, desk arrangement, and the young woman’s position behind the seated staff. Exclude the seated security employee and officer from the close framing; focus on the young woman’s tense face and tightly clasped hands.

REALIZE FIGURATIVE LANGUAGE AS A LIVE-ACTION SHOT: the Korean shot
text may describe characters metaphorically, figuratively or with
exaggeration. Photograph what a real movie camera would actually
record on a physical set — exaggerated or figurative impressions
become realistic staging choices (lighting, distance, angle,
wardrobe), not literal fantasy imagery.
EVERY CHARACTER IS A HUMAN BEING: unless the story explicitly
features non-human or virtual beings (as in science fiction or
fantasy), every character — however indirectly, vaguely or
figuratively the text describes them — IS a real human. When the
text gives no direct visual description of a person, IMAGINE one
and still show them as a concrete, fully-formed human being:
always render the human form to the maximum extent — build,
posture, face, hands, clothing — never reduce a person to a shape,
blob, solid silhouette or abstract mass.

EXPRESSIONS ARE ACTED, NEVER ANATOMICAL: when the text describes a
person's eyes, face or presence emotionally or figuratively (vacant,
hollow, dazed, burning, lifeless gaze and the like), it describes an
ACTOR'S PERFORMANCE captured by a real camera — realize it ONLY
through gaze direction, focus, eyelids, facial muscles, stillness
and posture. Every living person's eyes remain anatomically normal
human eyes with a natural iris and pupil, natural sclera and normal
proportions; NEVER whiten, blank out, cloud over, glow, enlarge or
otherwise alter eyeballs, skin or anatomy — unless the story
explicitly declares that being non-human or supernatural in form.

PROPS FACE THE RIGHT WAY: every handheld or used object must be
oriented exactly as its real-world use requires. A person reading,
watching or operating something (a phone, a photograph, a paper,
any device) has its functional side — screen, front, page — facing
THEIR OWN eyes; the camera then sees whatever side the staging
geometry implies (often its back). Show the functional side to the
camera ONLY when the shot text itself stages it toward the viewer.
Never flip, mirror or reverse an object's front and back.

CARRIED STATE (persist exactly — must match the neighbouring shots of this scene): The same section of CCTV footage remains abnormally dark and unidentifiable on every replay. Su-ri-young’s doll-adorned backpack remains with her.

PEOPLE: the SHOT TEXT alone decides whether any person is visible in this shot. IF a person appears, they must be one of: 수리영 (한국인 20대 초반 여성, 긴 검은 생머리, 젊은 타원형 얼굴) — never anyone else, and never add a person the shot text does not show. IF only part of a person is in frame (a hand, arm, foot, back, silhouette), that body part belongs to the specific person the shot text names — its sex, age, build, skin and grooming must unmistakably match that person's profile above.

No text, captions, watermarks or annotations anywhere.

VISUAL APPROACH — this take (camera, framing, composition):
Photograph the close-up with a compressed, observational lens feel, arranging her face and clasped hands as one compact vertical block in the center while retaining a thin layer of spatial context around her. Keep more facial contour and hand texture legible under the cool, even office illumination, with the active monitor only marginally readable as an indistinct dark plane aligned to her stare. Emphasize locked posture, a firm jaw, and unwavering eyeline rather than softness or glamour.

PRIORITY OF INSTRUCTIONS: the VISUAL APPROACH above governs camera,
framing and composition ONLY. Every other instruction in this prompt —
location, time of day, people and their identity, poses, props,
carried state, physical rules, and the no-text rule — is binding
exactly as written earlier. If the VISUAL APPROACH appears to conflict
with any of them, the earlier instruction wins.

Edit this photograph. Apply ONLY the corrections listed below,
each in place, changing nothing else:

CORRECTIONS:
- Remove the air conditioner, window with blinds, and the background screen, replacing them with the plain, empty wall seen in the reference still.
- Darken the visible CCTV monitor screen so that the footage appears abnormally dark and completely unidentifiable.

PRESERVE EVERYTHING ELSE EXACTLY as in the original photograph —
every person, pose, face, garment, object, set feature, lighting,
camera and framing stay identical. No text, captions or
watermarks anywhere.

Create ONE FINAL photorealistic live-action film still of the moment below — contemporary South Korea, 2026; all people are Korean unless stated. TIME OF DAY (lock): late morning.

SHOT TEXT (authoritative, Korean): 수리영이 입을 벌린 채 사색이 되어 굳어버린 얼굴 클로즈업

LOCATION (lock): Beside the unattended computer desk inside the daytime police office, facing the phone-location display. The shot takes place here — the attached PREVIOUS SHOT STILL shows this exact place.

CAMERA & FRAME (follow exactly — this is the composition authority for this still; the reference images supply identity and place, never the framing):
- CAMERA: At Suriyoung's eye level, complete the lateral arc into a frontal close position and hold her stricken face centrally, with her parted mouth and drained expression carrying the frame. Keep the monitor direction just off camera so her frozen eyeline remains continuous with the preceding reveal, while the lower frame leaves room for the later tilt toward her writing hand.
- FRAMING SCALE: close-up
- KEY BACKGROUND ELEMENTS: 경찰서 책상 구역 (Temporarily unoccupied except for Suriyoung) — Desk edges recede behind and below Suriyoung from the monitor-facing side; used as Soft peripheral anchor preserving the police-office setting behind her reaction; 살인사건 확대 사진들 (Scattered across 혜수의 책상) — Their image-bearing sides face upward on the desk, but their details fall outside the focal plane; used as Blurred background evidence connecting her shock to the earlier discovery.
Compose the frame exactly as specified above — camera angle, subject scale and screen placement. Keep true physical scale between people and background elements: fixtures and distant objects occupy only the small screen area their real size and distance dictate; never enlarge a background object into a foreground presence.

THIS SHOT CONTINUES THE PREVIOUS SHOT: everything the attached still established — the place, its fixed features and wear, each person's clothing and state — persists. Any person in it who cannot move stays PRECISELY as photographed (body, pose, contact points, held objects); only the camera changes.

REALIZE FIGURATIVE LANGUAGE AS A LIVE-ACTION SHOT: the Korean shot
text may describe characters metaphorically, figuratively or with
exaggeration. Photograph what a real movie camera would actually
record on a physical set — exaggerated or figurative impressions
become realistic staging choices (lighting, distance, angle,
wardrobe), not literal fantasy imagery.
EVERY CHARACTER IS A HUMAN BEING: unless the story explicitly
features non-human or virtual beings (as in science fiction or
fantasy), every character — however indirectly, vaguely or
figuratively the text describes them — IS a real human. When the
text gives no direct visual description of a person, IMAGINE one
and still show them as a concrete, fully-formed human being:
always render the human form to the maximum extent — build,
posture, face, hands, clothing — never reduce a person to a shape,
blob, solid silhouette or abstract mass.

EXPRESSIONS ARE ACTED, NEVER ANATOMICAL: when the text describes a
person's eyes, face or presence emotionally or figuratively (vacant,
hollow, dazed, burning, lifeless gaze and the like), it describes an
ACTOR'S PERFORMANCE captured by a real camera — realize it ONLY
through gaze direction, focus, eyelids, facial muscles, stillness
and posture. Every living person's eyes remain anatomically normal
human eyes with a natural iris and pupil, natural sclera and normal
proportions; NEVER whiten, blank out, cloud over, glow, enlarge or
otherwise alter eyeballs, skin or anatomy — unless the story
explicitly declares that being non-human or supernatural in form.

PROPS FACE THE RIGHT WAY: every handheld or used object must be
oriented exactly as its real-world use requires. A person reading,
watching or operating something (a phone, a photograph, a paper,
any device) has its functional side — screen, front, page — facing
THEIR OWN eyes; the camera then sees whatever side the staging
geometry implies (often its back). Show the functional side to the
camera ONLY when the shot text itself stages it toward the viewer.
Never flip, mirror or reverse an object's front and back.

CARRIED STATE (persist exactly — must match the neighbouring shots of this scene): The offshore coordinate result remains visible on Kim’s monitor, and the murder-scene photographs remain spread over Hye-su’s desk. Su-ri-young still has her doll-adorned backpack and phone.

PEOPLE: the SHOT TEXT alone decides whether any person is visible in this shot. IF a person appears, they must be one of: 수리영 (한국인 20대 초반 여성, 긴 검은 생머리, 젊은 타원형 얼굴) — never anyone else, and never add a person the shot text does not show. IF only part of a person is in frame (a hand, arm, foot, back, silhouette), that body part belongs to the specific person the shot text names — its sex, age, build, skin and grooming must unmistakably match that person's profile above.

No text, captions, watermarks or annotations anywhere.

Create ONE FINAL photorealistic live-action film still of the moment below — contemporary South Korea, 2026; all people are Korean unless stated. TIME OF DAY (lock): late morning.

SHOT TEXT (authoritative, Korean): 수리영이 입을 벌린 채 사색이 되어 굳어버린 얼굴 클로즈업

LOCATION (lock): Beside the unattended computer desk inside the daytime police office, facing the phone-location display. The shot takes place here — the attached PREVIOUS SHOT STILL shows this exact place.

CAMERA & FRAME (follow exactly — this is the composition authority for this still; the reference images supply identity and place, never the framing):
- CAMERA: At Suriyoung's eye level, complete the lateral arc into a frontal close position and hold her stricken face centrally, with her parted mouth and drained expression carrying the frame. Keep the monitor direction just off camera so her frozen eyeline remains continuous with the preceding reveal, while the lower frame leaves room for the later tilt toward her writing hand.
- FRAMING SCALE: close-up
- KEY BACKGROUND ELEMENTS: 경찰서 책상 구역 (Temporarily unoccupied except for Suriyoung) — Desk edges recede behind and below Suriyoung from the monitor-facing side; used as Soft peripheral anchor preserving the police-office setting behind her reaction; 살인사건 확대 사진들 (Scattered across 혜수의 책상) — Their image-bearing sides face upward on the desk, but their details fall outside the focal plane; used as Blurred background evidence connecting her shock to the earlier discovery.
Compose the frame exactly as specified above — camera angle, subject scale and screen placement. Keep true physical scale between people and background elements: fixtures and distant objects occupy only the small screen area their real size and distance dictate; never enlarge a background object into a foreground presence.

THIS SHOT CONTINUES THE PREVIOUS SHOT: everything the attached still established — the place, its fixed features and wear, each person's clothing and state — persists. Any person in it who cannot move stays PRECISELY as photographed (body, pose, contact points, held objects); only the camera changes.

REALIZE FIGURATIVE LANGUAGE AS A LIVE-ACTION SHOT: the Korean shot
text may describe characters metaphorically, figuratively or with
exaggeration. Photograph what a real movie camera would actually
record on a physical set — exaggerated or figurative impressions
become realistic staging choices (lighting, distance, angle,
wardrobe), not literal fantasy imagery.
EVERY CHARACTER IS A HUMAN BEING: unless the story explicitly
features non-human or virtual beings (as in science fiction or
fantasy), every character — however indirectly, vaguely or
figuratively the text describes them — IS a real human. When the
text gives no direct visual description of a person, IMAGINE one
and still show them as a concrete, fully-formed human being:
always render the human form to the maximum extent — build,
posture, face, hands, clothing — never reduce a person to a shape,
blob, solid silhouette or abstract mass.

EXPRESSIONS ARE ACTED, NEVER ANATOMICAL: when the text describes a
person's eyes, face or presence emotionally or figuratively (vacant,
hollow, dazed, burning, lifeless gaze and the like), it describes an
ACTOR'S PERFORMANCE captured by a real camera — realize it ONLY
through gaze direction, focus, eyelids, facial muscles, stillness
and posture. Every living person's eyes remain anatomically normal
human eyes with a natural iris and pupil, natural sclera and normal
proportions; NEVER whiten, blank out, cloud over, glow, enlarge or
otherwise alter eyeballs, skin or anatomy — unless the story
explicitly declares that being non-human or supernatural in form.

PROPS FACE THE RIGHT WAY: every handheld or used object must be
oriented exactly as its real-world use requires. A person reading,
watching or operating something (a phone, a photograph, a paper,
any device) has its functional side — screen, front, page — facing
THEIR OWN eyes; the camera then sees whatever side the staging
geometry implies (often its back). Show the functional side to the
camera ONLY when the shot text itself stages it toward the viewer.
Never flip, mirror or reverse an object's front and back.

CARRIED STATE (persist exactly — must match the neighbouring shots of this scene): The offshore coordinate result remains visible on Kim’s monitor, and the murder-scene photographs remain spread over Hye-su’s desk. Su-ri-young still has her doll-adorned backpack and phone.

PEOPLE: the SHOT TEXT alone decides whether any person is visible in this shot. IF a person appears, they must be one of: 수리영 (한국인 20대 초반 여성, 긴 검은 생머리, 젊은 타원형 얼굴) — never anyone else, and never add a person the shot text does not show. IF only part of a person is in frame (a hand, arm, foot, back, silhouette), that body part belongs to the specific person the shot text names — its sex, age, build, skin and grooming must unmistakably match that person's profile above.

No text, captions, watermarks or annotations anywhere.

VISUAL APPROACH — this take (camera, framing, composition):
Use a locked, frontal close-up with a compressed lens feel, isolating Suriyoung’s face against an extremely shallow field of focus. Keep her face centered and nearly motionless, her parted lips, fixed eyeline, and drained facial tension rendered with restrained realism; let the desk and upward-facing photographs dissolve into soft peripheral shapes. Shape the late-morning office light gently across her face with subdued contrast, preserving open space beneath her chin for the eventual downward tilt.

PRIORITY OF INSTRUCTIONS: the VISUAL APPROACH above governs camera,
framing and composition ONLY. Every other instruction in this prompt —
location, time of day, people and their identity, poses, props,
carried state, physical rules, and the no-text rule — is binding
exactly as written earlier. If the VISUAL APPROACH appears to conflict
with any of them, the earlier instruction wins.

Create ONE FINAL photorealistic live-action film still of the moment below — contemporary South Korea, 2026; all people are Korean unless stated. TIME OF DAY (lock): late morning.

SHOT TEXT (authoritative, Korean): 수리영이 입을 벌린 채 사색이 되어 굳어버린 얼굴 클로즈업

LOCATION (lock): Beside the unattended computer desk inside the daytime police office, facing the phone-location display. The shot takes place here — the attached PREVIOUS SHOT STILL shows this exact place.

CAMERA & FRAME (follow exactly — this is the composition authority for this still; the reference images supply identity and place, never the framing):
- CAMERA: At Suriyoung's eye level, complete the lateral arc into a frontal close position and hold her stricken face centrally, with her parted mouth and drained expression carrying the frame. Keep the monitor direction just off camera so her frozen eyeline remains continuous with the preceding reveal, while the lower frame leaves room for the later tilt toward her writing hand.
- FRAMING SCALE: close-up
- KEY BACKGROUND ELEMENTS: 경찰서 책상 구역 (Temporarily unoccupied except for Suriyoung) — Desk edges recede behind and below Suriyoung from the monitor-facing side; used as Soft peripheral anchor preserving the police-office setting behind her reaction; 살인사건 확대 사진들 (Scattered across 혜수의 책상) — Their image-bearing sides face upward on the desk, but their details fall outside the focal plane; used as Blurred background evidence connecting her shock to the earlier discovery.
Compose the frame exactly as specified above — camera angle, subject scale and screen placement. Keep true physical scale between people and background elements: fixtures and distant objects occupy only the small screen area their real size and distance dictate; never enlarge a background object into a foreground presence.

THIS SHOT CONTINUES THE PREVIOUS SHOT: everything the attached still established — the place, its fixed features and wear, each person's clothing and state — persists. Any person in it who cannot move stays PRECISELY as photographed (body, pose, contact points, held objects); only the camera changes.

REALIZE FIGURATIVE LANGUAGE AS A LIVE-ACTION SHOT: the Korean shot
text may describe characters metaphorically, figuratively or with
exaggeration. Photograph what a real movie camera would actually
record on a physical set — exaggerated or figurative impressions
become realistic staging choices (lighting, distance, angle,
wardrobe), not literal fantasy imagery.
EVERY CHARACTER IS A HUMAN BEING: unless the story explicitly
features non-human or virtual beings (as in science fiction or
fantasy), every character — however indirectly, vaguely or
figuratively the text describes them — IS a real human. When the
text gives no direct visual description of a person, IMAGINE one
and still show them as a concrete, fully-formed human being:
always render the human form to the maximum extent — build,
posture, face, hands, clothing — never reduce a person to a shape,
blob, solid silhouette or abstract mass.

EXPRESSIONS ARE ACTED, NEVER ANATOMICAL: when the text describes a
person's eyes, face or presence emotionally or figuratively (vacant,
hollow, dazed, burning, lifeless gaze and the like), it describes an
ACTOR'S PERFORMANCE captured by a real camera — realize it ONLY
through gaze direction, focus, eyelids, facial muscles, stillness
and posture. Every living person's eyes remain anatomically normal
human eyes with a natural iris and pupil, natural sclera and normal
proportions; NEVER whiten, blank out, cloud over, glow, enlarge or
otherwise alter eyeballs, skin or anatomy — unless the story
explicitly declares that being non-human or supernatural in form.

PROPS FACE THE RIGHT WAY: every handheld or used object must be
oriented exactly as its real-world use requires. A person reading,
watching or operating something (a phone, a photograph, a paper,
any device) has its functional side — screen, front, page — facing
THEIR OWN eyes; the camera then sees whatever side the staging
geometry implies (often its back). Show the functional side to the
camera ONLY when the shot text itself stages it toward the viewer.
Never flip, mirror or reverse an object's front and back.

CARRIED STATE (persist exactly — must match the neighbouring shots of this scene): The offshore coordinate result remains visible on Kim’s monitor, and the murder-scene photographs remain spread over Hye-su’s desk. Su-ri-young still has her doll-adorned backpack and phone.

PEOPLE: the SHOT TEXT alone decides whether any person is visible in this shot. IF a person appears, they must be one of: 수리영 (한국인 20대 초반 여성, 긴 검은 생머리, 젊은 타원형 얼굴) — never anyone else, and never add a person the shot text does not show. IF only part of a person is in frame (a hand, arm, foot, back, silhouette), that body part belongs to the specific person the shot text names — its sex, age, build, skin and grooming must unmistakably match that person's profile above.

No text, captions, watermarks or annotations anywhere.

VISUAL APPROACH — this take (camera, framing, composition):
Photograph the same frontal eye-level close-up with a slightly broader, more immediate lens feel, allowing the receding desk edges to create quiet diagonal depth around her centrally held face. Maintain enough environmental definition to register the police-office workspace while keeping the scattered photographs unreadably blurred, making her frozen stare and open mouth feel abruptly exposed rather than isolated. Use plain, even late-morning office illumination with natural facial texture and retain deliberate lower-frame breathing room.

PRIORITY OF INSTRUCTIONS: the VISUAL APPROACH above governs camera,
framing and composition ONLY. Every other instruction in this prompt —
location, time of day, people and their identity, poses, props,
carried state, physical rules, and the no-text rule — is binding
exactly as written earlier. If the VISUAL APPROACH appears to conflict
with any of them, the earlier instruction wins.

Edit this photograph. Apply ONLY the corrections listed below,
each in place, changing nothing else:

CORRECTIONS:
- Add the green cap from the character reference to her head.
- Add the dark blue backpack straps over both of her shoulders.
- Remove the black monitor behind her head so the background shows the open police office space.

PRESERVE EVERYTHING ELSE EXACTLY as in the original photograph —
every person, pose, face, garment, object, set feature, lighting,
camera and framing stay identical. No text, captions or
watermarks anywhere.

Create ONE FINAL photorealistic live-action film still of the moment below — contemporary South Korea, 2026; all people are Korean unless stated. TIME OF DAY (lock): late morning.

SHOT TEXT (authoritative, Korean): 수리영의 손이 접은 메모지를 자신의 바지 주머니에 반쯤 밀어 넣고 있는 동작의 찰나

LOCATION (lock): At the detective's workstation inside the daytime police office, where she secretly pockets the folded coordinate note. The shot takes place here — the attached PREVIOUS SHOT STILL shows this exact place.

CAMERA & FRAME (follow exactly — this is the composition authority for this still; the reference images supply identity and place, never the framing):
- CAMERA: At waist height beside Suriyoung, finish the downward tilt on the instant her hand pushes the folded note halfway into her trouser pocket. Isolate the hand, paper edge, and pocket opening while retaining a narrow strip of torso and desk in the background for physical scale; the hand rests near the lower center rather than letting the note overwhelm the frame.
- FRAMING SCALE: insert close-up on a detail
- KEY BACKGROUND ELEMENTS: 수리영의 바지 주머니 (Receiving the folded note) — The pocket opening angles toward the camera enough to show the note entering halfway; used as Primary concealment anchor beneath Suriyoung's hand; 접은 좌표 메모지 (Folded and halfway inserted) — A portion of the written face remains visible above the pocket opening while the rest is concealed; used as The written edge remains visible as the key narrative detail; 책상 가장자리 (In place) — The side edge crosses the upper background behind Suriyoung's waist; used as Soft background reference connecting the gesture to the police-office workstation.
Compose the frame exactly as specified above — camera angle, subject scale and screen placement. Keep true physical scale between people and background elements: fixtures and distant objects occupy only the small screen area their real size and distance dictate; never enlarge a background object into a foreground presence.

THIS SHOT CONTINUES THE PREVIOUS SHOT: everything the attached still established — the place, its fixed features and wear, each person's clothing and state — persists. Any person in it who cannot move stays PRECISELY as photographed (body, pose, contact points, held objects); only the camera changes.

REALIZE FIGURATIVE LANGUAGE AS A LIVE-ACTION SHOT: the Korean shot
text may describe characters metaphorically, figuratively or with
exaggeration. Photograph what a real movie camera would actually
record on a physical set — exaggerated or figurative impressions
become realistic staging choices (lighting, distance, angle,
wardrobe), not literal fantasy imagery.
EVERY CHARACTER IS A HUMAN BEING: unless the story explicitly
features non-human or virtual beings (as in science fiction or
fantasy), every character — however indirectly, vaguely or
figuratively the text describes them — IS a real human. When the
text gives no direct visual description of a person, IMAGINE one
and still show them as a concrete, fully-formed human being:
always render the human form to the maximum extent — build,
posture, face, hands, clothing — never reduce a person to a shape,
blob, solid silhouette or abstract mass.

EXPRESSIONS ARE ACTED, NEVER ANATOMICAL: when the text describes a
person's eyes, face or presence emotionally or figuratively (vacant,
hollow, dazed, burning, lifeless gaze and the like), it describes an
ACTOR'S PERFORMANCE captured by a real camera — realize it ONLY
through gaze direction, focus, eyelids, facial muscles, stillness
and posture. Every living person's eyes remain anatomically normal
human eyes with a natural iris and pupil, natural sclera and normal
proportions; NEVER whiten, blank out, cloud over, glow, enlarge or
otherwise alter eyeballs, skin or anatomy — unless the story
explicitly declares that being non-human or supernatural in form.

PROPS FACE THE RIGHT WAY: every handheld or used object must be
oriented exactly as its real-world use requires. A person reading,
watching or operating something (a phone, a photograph, a paper,
any device) has its functional side — screen, front, page — facing
THEIR OWN eyes; the camera then sees whatever side the staging
geometry implies (often its back). Show the functional side to the
camera ONLY when the shot text itself stages it toward the viewer.
Never flip, mirror or reverse an object's front and back.

CARRIED STATE (persist exactly — must match the neighbouring shots of this scene): Su-ri-young places the folded note containing Min-suk’s offshore coordinates into her pocket. Her doll-adorned backpack and phone remain with her.

PEOPLE: the SHOT TEXT alone decides whether any person is visible in this shot. IF a person appears, they must be one of: 수리영 (한국인 20대 초반 여성, 긴 검은 생머리, 젊은 타원형 얼굴) — never anyone else, and never add a person the shot text does not show. IF only part of a person is in frame (a hand, arm, foot, back, silhouette), that body part belongs to the specific person the shot text names — its sex, age, build, skin and grooming must unmistakably match that person's profile above.

No text, captions, watermarks or annotations anywhere.

Create ONE FINAL photorealistic live-action film still of the moment below — contemporary South Korea, 2026; all people are Korean unless stated. TIME OF DAY (lock): late morning.

SHOT TEXT (authoritative, Korean): 수리영의 손이 접은 메모지를 자신의 바지 주머니에 반쯤 밀어 넣고 있는 동작의 찰나

LOCATION (lock): At the detective's workstation inside the daytime police office, where she secretly pockets the folded coordinate note. The shot takes place here — the attached PREVIOUS SHOT STILL shows this exact place.

CAMERA & FRAME (follow exactly — this is the composition authority for this still; the reference images supply identity and place, never the framing):
- CAMERA: At waist height beside Suriyoung, finish the downward tilt on the instant her hand pushes the folded note halfway into her trouser pocket. Isolate the hand, paper edge, and pocket opening while retaining a narrow strip of torso and desk in the background for physical scale; the hand rests near the lower center rather than letting the note overwhelm the frame.
- FRAMING SCALE: insert close-up on a detail
- KEY BACKGROUND ELEMENTS: 수리영의 바지 주머니 (Receiving the folded note) — The pocket opening angles toward the camera enough to show the note entering halfway; used as Primary concealment anchor beneath Suriyoung's hand; 접은 좌표 메모지 (Folded and halfway inserted) — A portion of the written face remains visible above the pocket opening while the rest is concealed; used as The written edge remains visible as the key narrative detail; 책상 가장자리 (In place) — The side edge crosses the upper background behind Suriyoung's waist; used as Soft background reference connecting the gesture to the police-office workstation.
Compose the frame exactly as specified above — camera angle, subject scale and screen placement. Keep true physical scale between people and background elements: fixtures and distant objects occupy only the small screen area their real size and distance dictate; never enlarge a background object into a foreground presence.

THIS SHOT CONTINUES THE PREVIOUS SHOT: everything the attached still established — the place, its fixed features and wear, each person's clothing and state — persists. Any person in it who cannot move stays PRECISELY as photographed (body, pose, contact points, held objects); only the camera changes.

REALIZE FIGURATIVE LANGUAGE AS A LIVE-ACTION SHOT: the Korean shot
text may describe characters metaphorically, figuratively or with
exaggeration. Photograph what a real movie camera would actually
record on a physical set — exaggerated or figurative impressions
become realistic staging choices (lighting, distance, angle,
wardrobe), not literal fantasy imagery.
EVERY CHARACTER IS A HUMAN BEING: unless the story explicitly
features non-human or virtual beings (as in science fiction or
fantasy), every character — however indirectly, vaguely or
figuratively the text describes them — IS a real human. When the
text gives no direct visual description of a person, IMAGINE one
and still show them as a concrete, fully-formed human being:
always render the human form to the maximum extent — build,
posture, face, hands, clothing — never reduce a person to a shape,
blob, solid silhouette or abstract mass.

EXPRESSIONS ARE ACTED, NEVER ANATOMICAL: when the text describes a
person's eyes, face or presence emotionally or figuratively (vacant,
hollow, dazed, burning, lifeless gaze and the like), it describes an
ACTOR'S PERFORMANCE captured by a real camera — realize it ONLY
through gaze direction, focus, eyelids, facial muscles, stillness
and posture. Every living person's eyes remain anatomically normal
human eyes with a natural iris and pupil, natural sclera and normal
proportions; NEVER whiten, blank out, cloud over, glow, enlarge or
otherwise alter eyeballs, skin or anatomy — unless the story
explicitly declares that being non-human or supernatural in form.

PROPS FACE THE RIGHT WAY: every handheld or used object must be
oriented exactly as its real-world use requires. A person reading,
watching or operating something (a phone, a photograph, a paper,
any device) has its functional side — screen, front, page — facing
THEIR OWN eyes; the camera then sees whatever side the staging
geometry implies (often its back). Show the functional side to the
camera ONLY when the shot text itself stages it toward the viewer.
Never flip, mirror or reverse an object's front and back.

CARRIED STATE (persist exactly — must match the neighbouring shots of this scene): Su-ri-young places the folded note containing Min-suk’s offshore coordinates into her pocket. Her doll-adorned backpack and phone remain with her.

PEOPLE: the SHOT TEXT alone decides whether any person is visible in this shot. IF a person appears, they must be one of: 수리영 (한국인 20대 초반 여성, 긴 검은 생머리, 젊은 타원형 얼굴) — never anyone else, and never add a person the shot text does not show. IF only part of a person is in frame (a hand, arm, foot, back, silhouette), that body part belongs to the specific person the shot text names — its sex, age, build, skin and grooming must unmistakably match that person's profile above.

No text, captions, watermarks or annotations anywhere.

VISUAL APPROACH — this take (camera, framing, composition):
Photograph the insert with an intimate, slightly compressed lens feel and very shallow focus, holding crisp detail on the fingertips, exposed paper edge, and pocket seam while the torso strip and desk edge dissolve softly behind them. Use subdued late-morning office light falling gently across the hand, with restrained contrast and a slight falloff toward the frame edges to make the concealment feel furtive and immediate.

PRIORITY OF INSTRUCTIONS: the VISUAL APPROACH above governs camera,
framing and composition ONLY. Every other instruction in this prompt —
location, time of day, people and their identity, poses, props,
carried state, physical rules, and the no-text rule — is binding
exactly as written earlier. If the VISUAL APPROACH appears to conflict
with any of them, the earlier instruction wins.

Create ONE FINAL photorealistic live-action film still of the moment below — contemporary South Korea, 2026; all people are Korean unless stated. TIME OF DAY (lock): late morning.

SHOT TEXT (authoritative, Korean): 수리영의 손이 접은 메모지를 자신의 바지 주머니에 반쯤 밀어 넣고 있는 동작의 찰나

LOCATION (lock): At the detective's workstation inside the daytime police office, where she secretly pockets the folded coordinate note. The shot takes place here — the attached PREVIOUS SHOT STILL shows this exact place.

CAMERA & FRAME (follow exactly — this is the composition authority for this still; the reference images supply identity and place, never the framing):
- CAMERA: At waist height beside Suriyoung, finish the downward tilt on the instant her hand pushes the folded note halfway into her trouser pocket. Isolate the hand, paper edge, and pocket opening while retaining a narrow strip of torso and desk in the background for physical scale; the hand rests near the lower center rather than letting the note overwhelm the frame.
- FRAMING SCALE: insert close-up on a detail
- KEY BACKGROUND ELEMENTS: 수리영의 바지 주머니 (Receiving the folded note) — The pocket opening angles toward the camera enough to show the note entering halfway; used as Primary concealment anchor beneath Suriyoung's hand; 접은 좌표 메모지 (Folded and halfway inserted) — A portion of the written face remains visible above the pocket opening while the rest is concealed; used as The written edge remains visible as the key narrative detail; 책상 가장자리 (In place) — The side edge crosses the upper background behind Suriyoung's waist; used as Soft background reference connecting the gesture to the police-office workstation.
Compose the frame exactly as specified above — camera angle, subject scale and screen placement. Keep true physical scale between people and background elements: fixtures and distant objects occupy only the small screen area their real size and distance dictate; never enlarge a background object into a foreground presence.

THIS SHOT CONTINUES THE PREVIOUS SHOT: everything the attached still established — the place, its fixed features and wear, each person's clothing and state — persists. Any person in it who cannot move stays PRECISELY as photographed (body, pose, contact points, held objects); only the camera changes.

REALIZE FIGURATIVE LANGUAGE AS A LIVE-ACTION SHOT: the Korean shot
text may describe characters metaphorically, figuratively or with
exaggeration. Photograph what a real movie camera would actually
record on a physical set — exaggerated or figurative impressions
become realistic staging choices (lighting, distance, angle,
wardrobe), not literal fantasy imagery.
EVERY CHARACTER IS A HUMAN BEING: unless the story explicitly
features non-human or virtual beings (as in science fiction or
fantasy), every character — however indirectly, vaguely or
figuratively the text describes them — IS a real human. When the
text gives no direct visual description of a person, IMAGINE one
and still show them as a concrete, fully-formed human being:
always render the human form to the maximum extent — build,
posture, face, hands, clothing — never reduce a person to a shape,
blob, solid silhouette or abstract mass.

EXPRESSIONS ARE ACTED, NEVER ANATOMICAL: when the text describes a
person's eyes, face or presence emotionally or figuratively (vacant,
hollow, dazed, burning, lifeless gaze and the like), it describes an
ACTOR'S PERFORMANCE captured by a real camera — realize it ONLY
through gaze direction, focus, eyelids, facial muscles, stillness
and posture. Every living person's eyes remain anatomically normal
human eyes with a natural iris and pupil, natural sclera and normal
proportions; NEVER whiten, blank out, cloud over, glow, enlarge or
otherwise alter eyeballs, skin or anatomy — unless the story
explicitly declares that being non-human or supernatural in form.

PROPS FACE THE RIGHT WAY: every handheld or used object must be
oriented exactly as its real-world use requires. A person reading,
watching or operating something (a phone, a photograph, a paper,
any device) has its functional side — screen, front, page — facing
THEIR OWN eyes; the camera then sees whatever side the staging
geometry implies (often its back). Show the functional side to the
camera ONLY when the shot text itself stages it toward the viewer.
Never flip, mirror or reverse an object's front and back.

CARRIED STATE (persist exactly — must match the neighbouring shots of this scene): Su-ri-young places the folded note containing Min-suk’s offshore coordinates into her pocket. Her doll-adorned backpack and phone remain with her.

PEOPLE: the SHOT TEXT alone decides whether any person is visible in this shot. IF a person appears, they must be one of: 수리영 (한국인 20대 초반 여성, 긴 검은 생머리, 젊은 타원형 얼굴) — never anyone else, and never add a person the shot text does not show. IF only part of a person is in frame (a hand, arm, foot, back, silhouette), that body part belongs to the specific person the shot text names — its sex, age, build, skin and grooming must unmistakably match that person's profile above.

No text, captions, watermarks or annotations anywhere.

VISUAL APPROACH — this take (camera, framing, composition):
Use a natural, moderately wide lens feel with firmer depth, composing the hand and angled pocket opening along a clean diagonal toward the lower center while keeping the narrow torso strip and desk edge legible as spatial anchors. Favor even late-morning ambient office illumination with subtle directional modeling, crisp fabric and paper texture, and controlled neutral color so the halfway-completed gesture reads with procedural clarity rather than visual melodrama.

PRIORITY OF INSTRUCTIONS: the VISUAL APPROACH above governs camera,
framing and composition ONLY. Every other instruction in this prompt —
location, time of day, people and their identity, poses, props,
carried state, physical rules, and the no-text rule — is binding
exactly as written earlier. If the VISUAL APPROACH appears to conflict
with any of them, the earlier instruction wins.

Edit this photograph. Apply ONLY the corrections listed below,
each in place, changing nothing else:

CORRECTIONS:
- Add the blue backpack straps over the shoulders and chest of the beige hoodie.
- Remove the person sitting at the desk in the background so only Suriyoung remains.

PRESERVE EVERYTHING ELSE EXACTLY as in the original photograph —
every person, pose, face, garment, object, set feature, lighting,
camera and framing stay identical. No text, captions or
watermarks anywhere.

Create ONE FINAL photorealistic live-action film still of the moment below — contemporary South Korea, 2026; all people are Korean unless stated. TIME OF DAY (lock): late morning.

SHOT TEXT (authoritative, Korean): 수리영이 접은 종이 지도들을 열린 백팩 안으로 밀어 넣은 채 멈춘 손 클로즈업

LOCATION (lock): At the same computer booth inside the dark internet café, beside the desk and open backpack in the monitor glow. The shot takes place here — the attached PREVIOUS SHOT STILL shows this exact place.

CAMERA & FRAME (follow exactly — this is the composition authority for this still; the reference images supply identity and place, never the framing):
- CAMERA: From just above and slightly in front of the open backpack, hold a static downward insert on Suriyoung's paused hand after it pushes the folded maps inside. Keep the backpack opening below forty percent of the frame and include the desk edge and part of her forearm as scale references, with the visible map corners clustered at lower center.
- FRAMING SCALE: insert close-up on a detail
- KEY BACKGROUND ELEMENTS: 열린 백팩 (Open) — The opening faces upward and slightly toward the camera, revealing the newly inserted folded maps; used as Primary receiving space beneath Suriyoung's paused hand; 접은 종이 지도들 (Folded and inserted) — Folded printed faces and edges remain partially visible inside the backpack opening; used as Narrative detail visible within the bag without becoming oversized; PC방 책상 가장자리 (In use) — The near edge runs behind the backpack across the upper portion of frame; used as Surrounding scale reference framing the top of the insert.
Compose the frame exactly as specified above — camera angle, subject scale and screen placement. Keep true physical scale between people and background elements: fixtures and distant objects occupy only the small screen area their real size and distance dictate; never enlarge a background object into a foreground presence.

THIS SHOT CONTINUES THE PREVIOUS SHOT: everything the attached still established — the place, its fixed features and wear, each person's clothing and state — persists. Any person in it who cannot move stays PRECISELY as photographed (body, pose, contact points, held objects); only the camera changes.

PREVIOUS STILL USAGE (follow exactly — what to take from the attached still and what to exclude): Take the same dark PC-room desk, monitor glow, seating position, and the young woman’s continuing clothing and backpack. Exclude her face and the wider rows of computers; isolate her hands pushing the folded maps into the open backpack.

REALIZE FIGURATIVE LANGUAGE AS A LIVE-ACTION SHOT: the Korean shot
text may describe characters metaphorically, figuratively or with
exaggeration. Photograph what a real movie camera would actually
record on a physical set — exaggerated or figurative impressions
become realistic staging choices (lighting, distance, angle,
wardrobe), not literal fantasy imagery.
EVERY CHARACTER IS A HUMAN BEING: unless the story explicitly
features non-human or virtual beings (as in science fiction or
fantasy), every character — however indirectly, vaguely or
figuratively the text describes them — IS a real human. When the
text gives no direct visual description of a person, IMAGINE one
and still show them as a concrete, fully-formed human being:
always render the human form to the maximum extent — build,
posture, face, hands, clothing — never reduce a person to a shape,
blob, solid silhouette or abstract mass.

EXPRESSIONS ARE ACTED, NEVER ANATOMICAL: when the text describes a
person's eyes, face or presence emotionally or figuratively (vacant,
hollow, dazed, burning, lifeless gaze and the like), it describes an
ACTOR'S PERFORMANCE captured by a real camera — realize it ONLY
through gaze direction, focus, eyelids, facial muscles, stillness
and posture. Every living person's eyes remain anatomically normal
human eyes with a natural iris and pupil, natural sclera and normal
proportions; NEVER whiten, blank out, cloud over, glow, enlarge or
otherwise alter eyeballs, skin or anatomy — unless the story
explicitly declares that being non-human or supernatural in form.

PROPS FACE THE RIGHT WAY: every handheld or used object must be
oriented exactly as its real-world use requires. A person reading,
watching or operating something (a phone, a photograph, a paper,
any device) has its functional side — screen, front, page — facing
THEIR OWN eyes; the camera then sees whatever side the staging
geometry implies (often its back). Show the functional side to the
camera ONLY when the shot text itself stages it toward the viewer.
Never flip, mirror or reverse an object's front and back.

CARRIED STATE (persist exactly — must match the neighbouring shots of this scene): The printed maps are folded and placed inside Su-ri-young’s doll-adorned backpack; the original coordinate note remains in her possession.

PEOPLE: the SHOT TEXT alone decides whether any person is visible in this shot. IF a person appears, they must be one of: 수리영 (한국인 20대 초반 여성, 긴 검은 생머리, 젊은 타원형 얼굴) — never anyone else, and never add a person the shot text does not show. IF only part of a person is in frame (a hand, arm, foot, back, silhouette), that body part belongs to the specific person the shot text names — its sex, age, build, skin and grooming must unmistakably match that person's profile above.

No text, captions, watermarks or annotations anywhere.

Create ONE FINAL photorealistic live-action film still of the moment below — contemporary South Korea, 2026; all people are Korean unless stated. TIME OF DAY (lock): late morning.

SHOT TEXT (authoritative, Korean): 수리영이 접은 종이 지도들을 열린 백팩 안으로 밀어 넣은 채 멈춘 손 클로즈업

LOCATION (lock): At the same computer booth inside the dark internet café, beside the desk and open backpack in the monitor glow. The shot takes place here — the attached PREVIOUS SHOT STILL shows this exact place.

CAMERA & FRAME (follow exactly — this is the composition authority for this still; the reference images supply identity and place, never the framing):
- CAMERA: From just above and slightly in front of the open backpack, hold a static downward insert on Suriyoung's paused hand after it pushes the folded maps inside. Keep the backpack opening below forty percent of the frame and include the desk edge and part of her forearm as scale references, with the visible map corners clustered at lower center.
- FRAMING SCALE: insert close-up on a detail
- KEY BACKGROUND ELEMENTS: 열린 백팩 (Open) — The opening faces upward and slightly toward the camera, revealing the newly inserted folded maps; used as Primary receiving space beneath Suriyoung's paused hand; 접은 종이 지도들 (Folded and inserted) — Folded printed faces and edges remain partially visible inside the backpack opening; used as Narrative detail visible within the bag without becoming oversized; PC방 책상 가장자리 (In use) — The near edge runs behind the backpack across the upper portion of frame; used as Surrounding scale reference framing the top of the insert.
Compose the frame exactly as specified above — camera angle, subject scale and screen placement. Keep true physical scale between people and background elements: fixtures and distant objects occupy only the small screen area their real size and distance dictate; never enlarge a background object into a foreground presence.

THIS SHOT CONTINUES THE PREVIOUS SHOT: everything the attached still established — the place, its fixed features and wear, each person's clothing and state — persists. Any person in it who cannot move stays PRECISELY as photographed (body, pose, contact points, held objects); only the camera changes.

PREVIOUS STILL USAGE (follow exactly — what to take from the attached still and what to exclude): Take the same dark PC-room desk, monitor glow, seating position, and the young woman’s continuing clothing and backpack. Exclude her face and the wider rows of computers; isolate her hands pushing the folded maps into the open backpack.

REALIZE FIGURATIVE LANGUAGE AS A LIVE-ACTION SHOT: the Korean shot
text may describe characters metaphorically, figuratively or with
exaggeration. Photograph what a real movie camera would actually
record on a physical set — exaggerated or figurative impressions
become realistic staging choices (lighting, distance, angle,
wardrobe), not literal fantasy imagery.
EVERY CHARACTER IS A HUMAN BEING: unless the story explicitly
features non-human or virtual beings (as in science fiction or
fantasy), every character — however indirectly, vaguely or
figuratively the text describes them — IS a real human. When the
text gives no direct visual description of a person, IMAGINE one
and still show them as a concrete, fully-formed human being:
always render the human form to the maximum extent — build,
posture, face, hands, clothing — never reduce a person to a shape,
blob, solid silhouette or abstract mass.

EXPRESSIONS ARE ACTED, NEVER ANATOMICAL: when the text describes a
person's eyes, face or presence emotionally or figuratively (vacant,
hollow, dazed, burning, lifeless gaze and the like), it describes an
ACTOR'S PERFORMANCE captured by a real camera — realize it ONLY
through gaze direction, focus, eyelids, facial muscles, stillness
and posture. Every living person's eyes remain anatomically normal
human eyes with a natural iris and pupil, natural sclera and normal
proportions; NEVER whiten, blank out, cloud over, glow, enlarge or
otherwise alter eyeballs, skin or anatomy — unless the story
explicitly declares that being non-human or supernatural in form.

PROPS FACE THE RIGHT WAY: every handheld or used object must be
oriented exactly as its real-world use requires. A person reading,
watching or operating something (a phone, a photograph, a paper,
any device) has its functional side — screen, front, page — facing
THEIR OWN eyes; the camera then sees whatever side the staging
geometry implies (often its back). Show the functional side to the
camera ONLY when the shot text itself stages it toward the viewer.
Never flip, mirror or reverse an object's front and back.

CARRIED STATE (persist exactly — must match the neighbouring shots of this scene): The printed maps are folded and placed inside Su-ri-young’s doll-adorned backpack; the original coordinate note remains in her possession.

PEOPLE: the SHOT TEXT alone decides whether any person is visible in this shot. IF a person appears, they must be one of: 수리영 (한국인 20대 초반 여성, 긴 검은 생머리, 젊은 타원형 얼굴) — never anyone else, and never add a person the shot text does not show. IF only part of a person is in frame (a hand, arm, foot, back, silhouette), that body part belongs to the specific person the shot text names — its sex, age, build, skin and grooming must unmistakably match that person's profile above.

No text, captions, watermarks or annotations anywhere.

VISUAL APPROACH — this take (camera, framing, composition):
Use intimate, shallow-focus photography that makes the paused hand and clustered map corners the sharp dramatic center, with the forearm and desk edge falling gently softer. Let the monitor glow rake across the fingers and folded paper edges, revealing tactile skin, paper, and fabric texture while the backpack interior recedes into subdued shadow. Hold the composition tense and motionless, as if the action has stopped mid-thought.

PRIORITY OF INSTRUCTIONS: the VISUAL APPROACH above governs camera,
framing and composition ONLY. Every other instruction in this prompt —
location, time of day, people and their identity, poses, props,
carried state, physical rules, and the no-text rule — is binding
exactly as written earlier. If the VISUAL APPROACH appears to conflict
with any of them, the earlier instruction wins.

Create ONE FINAL photorealistic live-action film still of the moment below — contemporary South Korea, 2026; all people are Korean unless stated. TIME OF DAY (lock): late morning.

SHOT TEXT (authoritative, Korean): 수리영이 접은 종이 지도들을 열린 백팩 안으로 밀어 넣은 채 멈춘 손 클로즈업

LOCATION (lock): At the same computer booth inside the dark internet café, beside the desk and open backpack in the monitor glow. The shot takes place here — the attached PREVIOUS SHOT STILL shows this exact place.

CAMERA & FRAME (follow exactly — this is the composition authority for this still; the reference images supply identity and place, never the framing):
- CAMERA: From just above and slightly in front of the open backpack, hold a static downward insert on Suriyoung's paused hand after it pushes the folded maps inside. Keep the backpack opening below forty percent of the frame and include the desk edge and part of her forearm as scale references, with the visible map corners clustered at lower center.
- FRAMING SCALE: insert close-up on a detail
- KEY BACKGROUND ELEMENTS: 열린 백팩 (Open) — The opening faces upward and slightly toward the camera, revealing the newly inserted folded maps; used as Primary receiving space beneath Suriyoung's paused hand; 접은 종이 지도들 (Folded and inserted) — Folded printed faces and edges remain partially visible inside the backpack opening; used as Narrative detail visible within the bag without becoming oversized; PC방 책상 가장자리 (In use) — The near edge runs behind the backpack across the upper portion of frame; used as Surrounding scale reference framing the top of the insert.
Compose the frame exactly as specified above — camera angle, subject scale and screen placement. Keep true physical scale between people and background elements: fixtures and distant objects occupy only the small screen area their real size and distance dictate; never enlarge a background object into a foreground presence.

THIS SHOT CONTINUES THE PREVIOUS SHOT: everything the attached still established — the place, its fixed features and wear, each person's clothing and state — persists. Any person in it who cannot move stays PRECISELY as photographed (body, pose, contact points, held objects); only the camera changes.

PREVIOUS STILL USAGE (follow exactly — what to take from the attached still and what to exclude): Take the same dark PC-room desk, monitor glow, seating position, and the young woman’s continuing clothing and backpack. Exclude her face and the wider rows of computers; isolate her hands pushing the folded maps into the open backpack.

REALIZE FIGURATIVE LANGUAGE AS A LIVE-ACTION SHOT: the Korean shot
text may describe characters metaphorically, figuratively or with
exaggeration. Photograph what a real movie camera would actually
record on a physical set — exaggerated or figurative impressions
become realistic staging choices (lighting, distance, angle,
wardrobe), not literal fantasy imagery.
EVERY CHARACTER IS A HUMAN BEING: unless the story explicitly
features non-human or virtual beings (as in science fiction or
fantasy), every character — however indirectly, vaguely or
figuratively the text describes them — IS a real human. When the
text gives no direct visual description of a person, IMAGINE one
and still show them as a concrete, fully-formed human being:
always render the human form to the maximum extent — build,
posture, face, hands, clothing — never reduce a person to a shape,
blob, solid silhouette or abstract mass.

EXPRESSIONS ARE ACTED, NEVER ANATOMICAL: when the text describes a
person's eyes, face or presence emotionally or figuratively (vacant,
hollow, dazed, burning, lifeless gaze and the like), it describes an
ACTOR'S PERFORMANCE captured by a real camera — realize it ONLY
through gaze direction, focus, eyelids, facial muscles, stillness
and posture. Every living person's eyes remain anatomically normal
human eyes with a natural iris and pupil, natural sclera and normal
proportions; NEVER whiten, blank out, cloud over, glow, enlarge or
otherwise alter eyeballs, skin or anatomy — unless the story
explicitly declares that being non-human or supernatural in form.

PROPS FACE THE RIGHT WAY: every handheld or used object must be
oriented exactly as its real-world use requires. A person reading,
watching or operating something (a phone, a photograph, a paper,
any device) has its functional side — screen, front, page — facing
THEIR OWN eyes; the camera then sees whatever side the staging
geometry implies (often its back). Show the functional side to the
camera ONLY when the shot text itself stages it toward the viewer.
Never flip, mirror or reverse an object's front and back.

CARRIED STATE (persist exactly — must match the neighbouring shots of this scene): The printed maps are folded and placed inside Su-ri-young’s doll-adorned backpack; the original coordinate note remains in her possession.

PEOPLE: the SHOT TEXT alone decides whether any person is visible in this shot. IF a person appears, they must be one of: 수리영 (한국인 20대 초반 여성, 긴 검은 생머리, 젊은 타원형 얼굴) — never anyone else, and never add a person the shot text does not show. IF only part of a person is in frame (a hand, arm, foot, back, silhouette), that body part belongs to the specific person the shot text names — its sex, age, build, skin and grooming must unmistakably match that person's profile above.

No text, captions, watermarks or annotations anywhere.

VISUAL APPROACH — this take (camera, framing, composition):
Photograph the insert with restrained depth and crisp spatial clarity, organizing the desk edge as a firm upper boundary, the paused forearm as a diagonal lead, and the backpack opening with clustered map corners as the compact lower-center anchor. Keep the hand, folded paper, zipper rim, and desk edge legible across distinct depth layers rather than isolating a single fingertip. Use the existing monitor glow as soft, even directional illumination, giving the moment a cool, procedural stillness.

PRIORITY OF INSTRUCTIONS: the VISUAL APPROACH above governs camera,
framing and composition ONLY. Every other instruction in this prompt —
location, time of day, people and their identity, poses, props,
carried state, physical rules, and the no-text rule — is binding
exactly as written earlier. If the VISUAL APPROACH appears to conflict
with any of them, the earlier instruction wins.

Edit this photograph. Apply ONLY the corrections listed below,
each in place, changing nothing else:

CORRECTIONS:
- Scale down the backpack so its opening occupies less than forty percent of the frame, keeping the current camera angle.
- Reposition the hand downward so the fingers are physically pressing against the folded maps inside the backpack.

PRESERVE EVERYTHING ELSE EXACTLY as in the original photograph —
every person, pose, face, garment, object, set feature, lighting,
camera and framing stay identical. No text, captions or
watermarks anywhere.

Create ONE FINAL photorealistic live-action film still of the moment below — contemporary South Korea, 2026; all people are Korean unless stated. TIME OF DAY (lock): dusk.

SHOT TEXT (authoritative, Korean): 수리영의 굳게 쥔 주먹 사이로 구겨진 낡은 사진의 귀퉁이가 삐져나온 손 클로즈업

LOCATION (lock): Inside the ransacked rooftop apartment, near the daughter's bedroom as she prepares to leave at dusk. The shot takes place here — the attached PREVIOUS SHOT STILL shows this exact place.

CAMERA & FRAME (follow exactly — this is the composition authority for this still; the reference images supply identity and place, never the framing):
- CAMERA: At 수리영’s hand height, the camera tracks beside her toward the entrance in a tight close-up, with her clenched hand occupying the lower-center foreground and the crumpled photograph corner protruding between her fingers. The entrance remains ahead near the upper-right background while the disrupted living room passes softly behind, preserving the exit direction without enlarging the photograph unnaturally.
- FRAMING SCALE: close-up
- FRAME LAYOUT: 수리영's clenched hand in the lower-center of the frame, foreground, moves toward entrance; entrance in the upper-right of the frame, background.
- KEY BACKGROUND ELEMENTS: Old photograph (Crumpled and tightly held) — Only a crumpled corner of the image-bearing front is visible between 수리영's fingers; used as Small sharp detail protruding from the clenched hand; Scattered household objects (Disturbed and scattered) — Their displaced sides pass laterally behind the hand; used as Soft background context moving past during the hand-height track; Entrance (Visible ahead) — The interior-facing side lies ahead of 수리영's movement; used as Destination held ahead of the tracking hand.
Compose the frame exactly as specified above — camera angle, subject scale and screen placement. Keep true physical scale between people and background elements: fixtures and distant objects occupy only the small screen area their real size and distance dictate; never enlarge a background object into a foreground presence.

THIS SHOT CONTINUES THE PREVIOUS SHOT: everything the attached still established — the place, its fixed features and wear, each person's clothing and state — persists. Any person in it who cannot move stays PRECISELY as photographed (body, pose, contact points, held objects); only the camera changes.

PREVIOUS STILL USAGE (follow exactly — what to take from the attached still and what to exclude): Take the same dusk bedroom lighting, wall materials, large red circle, and the aged bloodstained photograph removed from its center. Exclude the wall and full red-circle composition from the close framing; show only the clenched hand gripping the crumpled photograph.

REALIZE FIGURATIVE LANGUAGE AS A LIVE-ACTION SHOT: the Korean shot
text may describe characters metaphorically, figuratively or with
exaggeration. Photograph what a real movie camera would actually
record on a physical set — exaggerated or figurative impressions
become realistic staging choices (lighting, distance, angle,
wardrobe), not literal fantasy imagery.
EVERY CHARACTER IS A HUMAN BEING: unless the story explicitly
features non-human or virtual beings (as in science fiction or
fantasy), every character — however indirectly, vaguely or
figuratively the text describes them — IS a real human. When the
text gives no direct visual description of a person, IMAGINE one
and still show them as a concrete, fully-formed human being:
always render the human form to the maximum extent — build,
posture, face, hands, clothing — never reduce a person to a shape,
blob, solid silhouette or abstract mass.

EXPRESSIONS ARE ACTED, NEVER ANATOMICAL: when the text describes a
person's eyes, face or presence emotionally or figuratively (vacant,
hollow, dazed, burning, lifeless gaze and the like), it describes an
ACTOR'S PERFORMANCE captured by a real camera — realize it ONLY
through gaze direction, focus, eyelids, facial muscles, stillness
and posture. Every living person's eyes remain anatomically normal
human eyes with a natural iris and pupil, natural sclera and normal
proportions; NEVER whiten, blank out, cloud over, glow, enlarge or
otherwise alter eyeballs, skin or anatomy — unless the story
explicitly declares that being non-human or supernatural in form.

PROPS FACE THE RIGHT WAY: every handheld or used object must be
oriented exactly as its real-world use requires. A person reading,
watching or operating something (a phone, a photograph, a paper,
any device) has its functional side — screen, front, page — facing
THEIR OWN eyes; the camera then sees whatever side the staging
geometry implies (often its back). Show the functional side to the
camera ONLY when the shot text itself stages it toward the viewer.
Never flip, mirror or reverse an object's front and back.

CARRIED STATE (persist exactly — must match the neighbouring shots of this scene): Su-ri-young has removed the old photograph from the wall and grips it tightly while wearing the backpack containing her maps. The large red circle remains on the bedroom wall, and the home remains ransacked.

PEOPLE: the SHOT TEXT alone decides whether any person is visible in this shot. IF a person appears, they must be one of: 수리영 (한국인 20대 초반 여성, 긴 검은 생머리, 젊은 타원형 얼굴) — never anyone else, and never add a person the shot text does not show. IF only part of a person is in frame (a hand, arm, foot, back, silhouette), that body part belongs to the specific person the shot text names — its sex, age, build, skin and grooming must unmistakably match that person's profile above.

No text, captions, watermarks or annotations anywhere.

Create ONE FINAL photorealistic live-action film still of the moment below — contemporary South Korea, 2026; all people are Korean unless stated. TIME OF DAY (lock): dusk.

SHOT TEXT (authoritative, Korean): 수리영의 굳게 쥔 주먹 사이로 구겨진 낡은 사진의 귀퉁이가 삐져나온 손 클로즈업

LOCATION (lock): Inside the ransacked rooftop apartment, near the daughter's bedroom as she prepares to leave at dusk. The shot takes place here — the attached PREVIOUS SHOT STILL shows this exact place.

CAMERA & FRAME (follow exactly — this is the composition authority for this still; the reference images supply identity and place, never the framing):
- CAMERA: At 수리영’s hand height, the camera tracks beside her toward the entrance in a tight close-up, with her clenched hand occupying the lower-center foreground and the crumpled photograph corner protruding between her fingers. The entrance remains ahead near the upper-right background while the disrupted living room passes softly behind, preserving the exit direction without enlarging the photograph unnaturally.
- FRAMING SCALE: close-up
- FRAME LAYOUT: 수리영's clenched hand in the lower-center of the frame, foreground, moves toward entrance; entrance in the upper-right of the frame, background.
- KEY BACKGROUND ELEMENTS: Old photograph (Crumpled and tightly held) — Only a crumpled corner of the image-bearing front is visible between 수리영's fingers; used as Small sharp detail protruding from the clenched hand; Scattered household objects (Disturbed and scattered) — Their displaced sides pass laterally behind the hand; used as Soft background context moving past during the hand-height track; Entrance (Visible ahead) — The interior-facing side lies ahead of 수리영's movement; used as Destination held ahead of the tracking hand.
Compose the frame exactly as specified above — camera angle, subject scale and screen placement. Keep true physical scale between people and background elements: fixtures and distant objects occupy only the small screen area their real size and distance dictate; never enlarge a background object into a foreground presence.

THIS SHOT CONTINUES THE PREVIOUS SHOT: everything the attached still established — the place, its fixed features and wear, each person's clothing and state — persists. Any person in it who cannot move stays PRECISELY as photographed (body, pose, contact points, held objects); only the camera changes.

PREVIOUS STILL USAGE (follow exactly — what to take from the attached still and what to exclude): Take the same dusk bedroom lighting, wall materials, large red circle, and the aged bloodstained photograph removed from its center. Exclude the wall and full red-circle composition from the close framing; show only the clenched hand gripping the crumpled photograph.

REALIZE FIGURATIVE LANGUAGE AS A LIVE-ACTION SHOT: the Korean shot
text may describe characters metaphorically, figuratively or with
exaggeration. Photograph what a real movie camera would actually
record on a physical set — exaggerated or figurative impressions
become realistic staging choices (lighting, distance, angle,
wardrobe), not literal fantasy imagery.
EVERY CHARACTER IS A HUMAN BEING: unless the story explicitly
features non-human or virtual beings (as in science fiction or
fantasy), every character — however indirectly, vaguely or
figuratively the text describes them — IS a real human. When the
text gives no direct visual description of a person, IMAGINE one
and still show them as a concrete, fully-formed human being:
always render the human form to the maximum extent — build,
posture, face, hands, clothing — never reduce a person to a shape,
blob, solid silhouette or abstract mass.

EXPRESSIONS ARE ACTED, NEVER ANATOMICAL: when the text describes a
person's eyes, face or presence emotionally or figuratively (vacant,
hollow, dazed, burning, lifeless gaze and the like), it describes an
ACTOR'S PERFORMANCE captured by a real camera — realize it ONLY
through gaze direction, focus, eyelids, facial muscles, stillness
and posture. Every living person's eyes remain anatomically normal
human eyes with a natural iris and pupil, natural sclera and normal
proportions; NEVER whiten, blank out, cloud over, glow, enlarge or
otherwise alter eyeballs, skin or anatomy — unless the story
explicitly declares that being non-human or supernatural in form.

PROPS FACE THE RIGHT WAY: every handheld or used object must be
oriented exactly as its real-world use requires. A person reading,
watching or operating something (a phone, a photograph, a paper,
any device) has its functional side — screen, front, page — facing
THEIR OWN eyes; the camera then sees whatever side the staging
geometry implies (often its back). Show the functional side to the
camera ONLY when the shot text itself stages it toward the viewer.
Never flip, mirror or reverse an object's front and back.

CARRIED STATE (persist exactly — must match the neighbouring shots of this scene): Su-ri-young has removed the old photograph from the wall and grips it tightly while wearing the backpack containing her maps. The large red circle remains on the bedroom wall, and the home remains ransacked.

PEOPLE: the SHOT TEXT alone decides whether any person is visible in this shot. IF a person appears, they must be one of: 수리영 (한국인 20대 초반 여성, 긴 검은 생머리, 젊은 타원형 얼굴) — never anyone else, and never add a person the shot text does not show. IF only part of a person is in frame (a hand, arm, foot, back, silhouette), that body part belongs to the specific person the shot text names — its sex, age, build, skin and grooming must unmistakably match that person's profile above.

No text, captions, watermarks or annotations anywhere.

VISUAL APPROACH — this take (camera, framing, composition):
Use an intimate, shallow-focus tracking image with the clenched hand and the protruding image-bearing corner rendered in tactile, unforgiving sharpness. Let the displaced household objects drift laterally as soft motion and tonal smears, while the entrance remains a subdued but readable destination in the upper-right. Shape the dusk illumination as restrained side light grazing the knuckles, crushed paper folds, and aged stains, with a tense, low-key falloff.

PRIORITY OF INSTRUCTIONS: the VISUAL APPROACH above governs camera,
framing and composition ONLY. Every other instruction in this prompt —
location, time of day, people and their identity, poses, props,
carried state, physical rules, and the no-text rule — is binding
exactly as written earlier. If the VISUAL APPROACH appears to conflict
with any of them, the earlier instruction wins.

Create ONE FINAL photorealistic live-action film still of the moment below — contemporary South Korea, 2026; all people are Korean unless stated. TIME OF DAY (lock): dusk.

SHOT TEXT (authoritative, Korean): 수리영의 굳게 쥔 주먹 사이로 구겨진 낡은 사진의 귀퉁이가 삐져나온 손 클로즈업

LOCATION (lock): Inside the ransacked rooftop apartment, near the daughter's bedroom as she prepares to leave at dusk. The shot takes place here — the attached PREVIOUS SHOT STILL shows this exact place.

CAMERA & FRAME (follow exactly — this is the composition authority for this still; the reference images supply identity and place, never the framing):
- CAMERA: At 수리영’s hand height, the camera tracks beside her toward the entrance in a tight close-up, with her clenched hand occupying the lower-center foreground and the crumpled photograph corner protruding between her fingers. The entrance remains ahead near the upper-right background while the disrupted living room passes softly behind, preserving the exit direction without enlarging the photograph unnaturally.
- FRAMING SCALE: close-up
- FRAME LAYOUT: 수리영's clenched hand in the lower-center of the frame, foreground, moves toward entrance; entrance in the upper-right of the frame, background.
- KEY BACKGROUND ELEMENTS: Old photograph (Crumpled and tightly held) — Only a crumpled corner of the image-bearing front is visible between 수리영's fingers; used as Small sharp detail protruding from the clenched hand; Scattered household objects (Disturbed and scattered) — Their displaced sides pass laterally behind the hand; used as Soft background context moving past during the hand-height track; Entrance (Visible ahead) — The interior-facing side lies ahead of 수리영's movement; used as Destination held ahead of the tracking hand.
Compose the frame exactly as specified above — camera angle, subject scale and screen placement. Keep true physical scale between people and background elements: fixtures and distant objects occupy only the small screen area their real size and distance dictate; never enlarge a background object into a foreground presence.

THIS SHOT CONTINUES THE PREVIOUS SHOT: everything the attached still established — the place, its fixed features and wear, each person's clothing and state — persists. Any person in it who cannot move stays PRECISELY as photographed (body, pose, contact points, held objects); only the camera changes.

PREVIOUS STILL USAGE (follow exactly — what to take from the attached still and what to exclude): Take the same dusk bedroom lighting, wall materials, large red circle, and the aged bloodstained photograph removed from its center. Exclude the wall and full red-circle composition from the close framing; show only the clenched hand gripping the crumpled photograph.

REALIZE FIGURATIVE LANGUAGE AS A LIVE-ACTION SHOT: the Korean shot
text may describe characters metaphorically, figuratively or with
exaggeration. Photograph what a real movie camera would actually
record on a physical set — exaggerated or figurative impressions
become realistic staging choices (lighting, distance, angle,
wardrobe), not literal fantasy imagery.
EVERY CHARACTER IS A HUMAN BEING: unless the story explicitly
features non-human or virtual beings (as in science fiction or
fantasy), every character — however indirectly, vaguely or
figuratively the text describes them — IS a real human. When the
text gives no direct visual description of a person, IMAGINE one
and still show them as a concrete, fully-formed human being:
always render the human form to the maximum extent — build,
posture, face, hands, clothing — never reduce a person to a shape,
blob, solid silhouette or abstract mass.

EXPRESSIONS ARE ACTED, NEVER ANATOMICAL: when the text describes a
person's eyes, face or presence emotionally or figuratively (vacant,
hollow, dazed, burning, lifeless gaze and the like), it describes an
ACTOR'S PERFORMANCE captured by a real camera — realize it ONLY
through gaze direction, focus, eyelids, facial muscles, stillness
and posture. Every living person's eyes remain anatomically normal
human eyes with a natural iris and pupil, natural sclera and normal
proportions; NEVER whiten, blank out, cloud over, glow, enlarge or
otherwise alter eyeballs, skin or anatomy — unless the story
explicitly declares that being non-human or supernatural in form.

PROPS FACE THE RIGHT WAY: every handheld or used object must be
oriented exactly as its real-world use requires. A person reading,
watching or operating something (a phone, a photograph, a paper,
any device) has its functional side — screen, front, page — facing
THEIR OWN eyes; the camera then sees whatever side the staging
geometry implies (often its back). Show the functional side to the
camera ONLY when the shot text itself stages it toward the viewer.
Never flip, mirror or reverse an object's front and back.

CARRIED STATE (persist exactly — must match the neighbouring shots of this scene): Su-ri-young has removed the old photograph from the wall and grips it tightly while wearing the backpack containing her maps. The large red circle remains on the bedroom wall, and the home remains ransacked.

PEOPLE: the SHOT TEXT alone decides whether any person is visible in this shot. IF a person appears, they must be one of: 수리영 (한국인 20대 초반 여성, 긴 검은 생머리, 젊은 타원형 얼굴) — never anyone else, and never add a person the shot text does not show. IF only part of a person is in frame (a hand, arm, foot, back, silhouette), that body part belongs to the specific person the shot text names — its sex, age, build, skin and grooming must unmistakably match that person's profile above.

No text, captions, watermarks or annotations anywhere.

VISUAL APPROACH — this take (camera, framing, composition):
Photograph the track with a steadier, more observational lens feel and moderately deep focus, preserving clear spatial progression from the lower-center hand to the upper-right entrance. Keep the fist dominant through proximity rather than distortion, while the ransacked room remains softly legible in layered depth behind it. Use the cool, fading dusk ambience with gentle interior contrast, allowing the photograph corner to register as a small, precise interruption in an otherwise controlled composition.

PRIORITY OF INSTRUCTIONS: the VISUAL APPROACH above governs camera,
framing and composition ONLY. Every other instruction in this prompt —
location, time of day, people and their identity, poses, props,
carried state, physical rules, and the no-text rule — is binding
exactly as written earlier. If the VISUAL APPROACH appears to conflict
with any of them, the earlier instruction wins.

Edit this photograph. Apply ONLY the corrections listed below,
each in place, changing nothing else:

CORRECTIONS:
- Remove the large sections of the photograph extending above and below the fist, leaving only a small, crumpled corner protruding from between the clenched fingers.

PRESERVE EVERYTHING ELSE EXACTLY as in the original photograph —
every person, pose, face, garment, object, set feature, lighting,
camera and framing stay identical. No text, captions or
watermarks anywhere.

Create ONE FINAL photorealistic live-action film still of the moment below — contemporary South Korea, 2026; all people are Korean unless stated. TIME OF DAY (lock): dusk, foggy.

SHOT TEXT (authoritative, Korean): 수리영이 인우의 등 뒤에 꼿꼿이 서서 그를 응시하는 측면

LOCATION (lock): On the foggy harbor landing beside the moored boat, where the boatman secures its line. The shot takes place here — the attached PREVIOUS SHOT STILL shows this exact place.

CAMERA & FRAME (follow exactly — this is the composition authority for this still; the reference images supply identity and place, never the framing):
- CAMERA: At chest height and perpendicular to their shared axis, the static medium frame keeps 인우’s back large in the left foreground while 수리영 appears in clear profile behind him toward the right midground. Their lateral separation remains readable against the boat and harbor, with the camera withholding a frontal view of either face.
- FRAMING SCALE: medium shot
- FRAME LAYOUT: 인우 in the middle-left of the frame, foreground; 수리영 in the middle-right of the frame, midground, looks toward 인우.
- KEY BACKGROUND ELEMENTS: Fishing boat (Arrived at the harbor) — Its docking side is visible behind 인우 and 수리영; used as Shared spatial anchor beside both characters; Mooring line (Being secured); used as Action anchor in 인우's nearer plane; Empty harbor (No other boat visible); used as Sparse background that reinforces the lack of alternatives; Fog (Dense); used as Softens and conceals the distant harbor.
Compose the frame exactly as specified above — camera angle, subject scale and screen placement. Keep true physical scale between people and background elements: fixtures and distant objects occupy only the small screen area their real size and distance dictate; never enlarge a background object into a foreground presence.

THIS SHOT CONTINUES THE PREVIOUS SHOT: everything the attached still established — the place, its fixed features and wear, each person's clothing and state — persists. Any person in it who cannot move stays PRECISELY as photographed (body, pose, contact points, held objects); only the camera changes.

REALIZE FIGURATIVE LANGUAGE AS A LIVE-ACTION SHOT: the Korean shot
text may describe characters metaphorically, figuratively or with
exaggeration. Photograph what a real movie camera would actually
record on a physical set — exaggerated or figurative impressions
become realistic staging choices (lighting, distance, angle,
wardrobe), not literal fantasy imagery.
EVERY CHARACTER IS A HUMAN BEING: unless the story explicitly
features non-human or virtual beings (as in science fiction or
fantasy), every character — however indirectly, vaguely or
figuratively the text describes them — IS a real human. When the
text gives no direct visual description of a person, IMAGINE one
and still show them as a concrete, fully-formed human being:
always render the human form to the maximum extent — build,
posture, face, hands, clothing — never reduce a person to a shape,
blob, solid silhouette or abstract mass.

EXPRESSIONS ARE ACTED, NEVER ANATOMICAL: when the text describes a
person's eyes, face or presence emotionally or figuratively (vacant,
hollow, dazed, burning, lifeless gaze and the like), it describes an
ACTOR'S PERFORMANCE captured by a real camera — realize it ONLY
through gaze direction, focus, eyelids, facial muscles, stillness
and posture. Every living person's eyes remain anatomically normal
human eyes with a natural iris and pupil, natural sclera and normal
proportions; NEVER whiten, blank out, cloud over, glow, enlarge or
otherwise alter eyeballs, skin or anatomy — unless the story
explicitly declares that being non-human or supernatural in form.

PROPS FACE THE RIGHT WAY: every handheld or used object must be
oriented exactly as its real-world use requires. A person reading,
watching or operating something (a phone, a photograph, a paper,
any device) has its functional side — screen, front, page — facing
THEIR OWN eyes; the camera then sees whatever side the staging
geometry implies (often its back). Show the functional side to the
camera ONLY when the shot text itself stages it toward the viewer.
Never flip, mirror or reverse an object's front and back.

CARRIED STATE (persist exactly — must match the neighbouring shots of this scene): Su-ri-young keeps the old photograph and her map-filled backpack with her beside In-woo’s boat. Her phone remains in her possession for the message to Hye-su.

PEOPLE: the SHOT TEXT alone decides whether any person is visible in this shot. IF a person appears, they must be one of: 수리영 (한국인 20대 초반 여성, 긴 검은 생머리, 젊은 타원형 얼굴); 인우 (한국인 20대 중반 남성, 짧은 검은 머리, 반듯한 얼굴형, 젊은 인상) — never anyone else, and never add a person the shot text does not show. IF only part of a person is in frame (a hand, arm, foot, back, silhouette), that body part belongs to the specific person the shot text names — its sex, age, build, skin and grooming must unmistakably match that person's profile above.

No text, captions, watermarks or annotations anywhere.

Create ONE FINAL photorealistic live-action film still of the moment below — contemporary South Korea, 2026; all people are Korean unless stated. TIME OF DAY (lock): dusk, foggy.

SHOT TEXT (authoritative, Korean): 수리영이 인우의 등 뒤에 꼿꼿이 서서 그를 응시하는 측면

LOCATION (lock): On the foggy harbor landing beside the moored boat, where the boatman secures its line. The shot takes place here — the attached PREVIOUS SHOT STILL shows this exact place.

CAMERA & FRAME (follow exactly — this is the composition authority for this still; the reference images supply identity and place, never the framing):
- CAMERA: At chest height and perpendicular to their shared axis, the static medium frame keeps 인우’s back large in the left foreground while 수리영 appears in clear profile behind him toward the right midground. Their lateral separation remains readable against the boat and harbor, with the camera withholding a frontal view of either face.
- FRAMING SCALE: medium shot
- FRAME LAYOUT: 인우 in the middle-left of the frame, foreground; 수리영 in the middle-right of the frame, midground, looks toward 인우.
- KEY BACKGROUND ELEMENTS: Fishing boat (Arrived at the harbor) — Its docking side is visible behind 인우 and 수리영; used as Shared spatial anchor beside both characters; Mooring line (Being secured); used as Action anchor in 인우's nearer plane; Empty harbor (No other boat visible); used as Sparse background that reinforces the lack of alternatives; Fog (Dense); used as Softens and conceals the distant harbor.
Compose the frame exactly as specified above — camera angle, subject scale and screen placement. Keep true physical scale between people and background elements: fixtures and distant objects occupy only the small screen area their real size and distance dictate; never enlarge a background object into a foreground presence.

THIS SHOT CONTINUES THE PREVIOUS SHOT: everything the attached still established — the place, its fixed features and wear, each person's clothing and state — persists. Any person in it who cannot move stays PRECISELY as photographed (body, pose, contact points, held objects); only the camera changes.

REALIZE FIGURATIVE LANGUAGE AS A LIVE-ACTION SHOT: the Korean shot
text may describe characters metaphorically, figuratively or with
exaggeration. Photograph what a real movie camera would actually
record on a physical set — exaggerated or figurative impressions
become realistic staging choices (lighting, distance, angle,
wardrobe), not literal fantasy imagery.
EVERY CHARACTER IS A HUMAN BEING: unless the story explicitly
features non-human or virtual beings (as in science fiction or
fantasy), every character — however indirectly, vaguely or
figuratively the text describes them — IS a real human. When the
text gives no direct visual description of a person, IMAGINE one
and still show them as a concrete, fully-formed human being:
always render the human form to the maximum extent — build,
posture, face, hands, clothing — never reduce a person to a shape,
blob, solid silhouette or abstract mass.

EXPRESSIONS ARE ACTED, NEVER ANATOMICAL: when the text describes a
person's eyes, face or presence emotionally or figuratively (vacant,
hollow, dazed, burning, lifeless gaze and the like), it describes an
ACTOR'S PERFORMANCE captured by a real camera — realize it ONLY
through gaze direction, focus, eyelids, facial muscles, stillness
and posture. Every living person's eyes remain anatomically normal
human eyes with a natural iris and pupil, natural sclera and normal
proportions; NEVER whiten, blank out, cloud over, glow, enlarge or
otherwise alter eyeballs, skin or anatomy — unless the story
explicitly declares that being non-human or supernatural in form.

PROPS FACE THE RIGHT WAY: every handheld or used object must be
oriented exactly as its real-world use requires. A person reading,
watching or operating something (a phone, a photograph, a paper,
any device) has its functional side — screen, front, page — facing
THEIR OWN eyes; the camera then sees whatever side the staging
geometry implies (often its back). Show the functional side to the
camera ONLY when the shot text itself stages it toward the viewer.
Never flip, mirror or reverse an object's front and back.

CARRIED STATE (persist exactly — must match the neighbouring shots of this scene): Su-ri-young keeps the old photograph and her map-filled backpack with her beside In-woo’s boat. Her phone remains in her possession for the message to Hye-su.

PEOPLE: the SHOT TEXT alone decides whether any person is visible in this shot. IF a person appears, they must be one of: 수리영 (한국인 20대 초반 여성, 긴 검은 생머리, 젊은 타원형 얼굴); 인우 (한국인 20대 중반 남성, 짧은 검은 머리, 반듯한 얼굴형, 젊은 인상) — never anyone else, and never add a person the shot text does not show. IF only part of a person is in frame (a hand, arm, foot, back, silhouette), that body part belongs to the specific person the shot text names — its sex, age, build, skin and grooming must unmistakably match that person's profile above.

No text, captions, watermarks or annotations anywhere.

VISUAL APPROACH — this take (camera, framing, composition):
Use a gently compressed lens feel and selective focus, holding Su-ri-young’s rigid profile and fixed gaze in crisp detail while In-woo’s large foreground back falls subtly soft. Let the cool, diffuse dusk filter through the dense fog, with restrained contrast and delicate edge separation from the boat, creating an intimate, suspended tension.

PRIORITY OF INSTRUCTIONS: the VISUAL APPROACH above governs camera,
framing and composition ONLY. Every other instruction in this prompt —
location, time of day, people and their identity, poses, props,
carried state, physical rules, and the no-text rule — is binding
exactly as written earlier. If the VISUAL APPROACH appears to conflict
with any of them, the earlier instruction wins.

Create ONE FINAL photorealistic live-action film still of the moment below — contemporary South Korea, 2026; all people are Korean unless stated. TIME OF DAY (lock): dusk, foggy.

SHOT TEXT (authoritative, Korean): 수리영이 인우의 등 뒤에 꼿꼿이 서서 그를 응시하는 측면

LOCATION (lock): On the foggy harbor landing beside the moored boat, where the boatman secures its line. The shot takes place here — the attached PREVIOUS SHOT STILL shows this exact place.

CAMERA & FRAME (follow exactly — this is the composition authority for this still; the reference images supply identity and place, never the framing):
- CAMERA: At chest height and perpendicular to their shared axis, the static medium frame keeps 인우’s back large in the left foreground while 수리영 appears in clear profile behind him toward the right midground. Their lateral separation remains readable against the boat and harbor, with the camera withholding a frontal view of either face.
- FRAMING SCALE: medium shot
- FRAME LAYOUT: 인우 in the middle-left of the frame, foreground; 수리영 in the middle-right of the frame, midground, looks toward 인우.
- KEY BACKGROUND ELEMENTS: Fishing boat (Arrived at the harbor) — Its docking side is visible behind 인우 and 수리영; used as Shared spatial anchor beside both characters; Mooring line (Being secured); used as Action anchor in 인우's nearer plane; Empty harbor (No other boat visible); used as Sparse background that reinforces the lack of alternatives; Fog (Dense); used as Softens and conceals the distant harbor.
Compose the frame exactly as specified above — camera angle, subject scale and screen placement. Keep true physical scale between people and background elements: fixtures and distant objects occupy only the small screen area their real size and distance dictate; never enlarge a background object into a foreground presence.

THIS SHOT CONTINUES THE PREVIOUS SHOT: everything the attached still established — the place, its fixed features and wear, each person's clothing and state — persists. Any person in it who cannot move stays PRECISELY as photographed (body, pose, contact points, held objects); only the camera changes.

REALIZE FIGURATIVE LANGUAGE AS A LIVE-ACTION SHOT: the Korean shot
text may describe characters metaphorically, figuratively or with
exaggeration. Photograph what a real movie camera would actually
record on a physical set — exaggerated or figurative impressions
become realistic staging choices (lighting, distance, angle,
wardrobe), not literal fantasy imagery.
EVERY CHARACTER IS A HUMAN BEING: unless the story explicitly
features non-human or virtual beings (as in science fiction or
fantasy), every character — however indirectly, vaguely or
figuratively the text describes them — IS a real human. When the
text gives no direct visual description of a person, IMAGINE one
and still show them as a concrete, fully-formed human being:
always render the human form to the maximum extent — build,
posture, face, hands, clothing — never reduce a person to a shape,
blob, solid silhouette or abstract mass.

EXPRESSIONS ARE ACTED, NEVER ANATOMICAL: when the text describes a
person's eyes, face or presence emotionally or figuratively (vacant,
hollow, dazed, burning, lifeless gaze and the like), it describes an
ACTOR'S PERFORMANCE captured by a real camera — realize it ONLY
through gaze direction, focus, eyelids, facial muscles, stillness
and posture. Every living person's eyes remain anatomically normal
human eyes with a natural iris and pupil, natural sclera and normal
proportions; NEVER whiten, blank out, cloud over, glow, enlarge or
otherwise alter eyeballs, skin or anatomy — unless the story
explicitly declares that being non-human or supernatural in form.

PROPS FACE THE RIGHT WAY: every handheld or used object must be
oriented exactly as its real-world use requires. A person reading,
watching or operating something (a phone, a photograph, a paper,
any device) has its functional side — screen, front, page — facing
THEIR OWN eyes; the camera then sees whatever side the staging
geometry implies (often its back). Show the functional side to the
camera ONLY when the shot text itself stages it toward the viewer.
Never flip, mirror or reverse an object's front and back.

CARRIED STATE (persist exactly — must match the neighbouring shots of this scene): Su-ri-young keeps the old photograph and her map-filled backpack with her beside In-woo’s boat. Her phone remains in her possession for the message to Hye-su.

PEOPLE: the SHOT TEXT alone decides whether any person is visible in this shot. IF a person appears, they must be one of: 수리영 (한국인 20대 초반 여성, 긴 검은 생머리, 젊은 타원형 얼굴); 인우 (한국인 20대 중반 남성, 짧은 검은 머리, 반듯한 얼굴형, 젊은 인상) — never anyone else, and never add a person the shot text does not show. IF only part of a person is in frame (a hand, arm, foot, back, silhouette), that body part belongs to the specific person the shot text names — its sex, age, build, skin and grooming must unmistakably match that person's profile above.

No text, captions, watermarks or annotations anywhere.

VISUAL APPROACH — this take (camera, framing, composition):
Use a natural, spatially open lens feel with deep focus, keeping In-woo’s foreground back, Su-ri-young’s profile, the mooring line, and the docking side of the boat clearly legible in layered depth. Favor a level, observational composition and broad soft dusk illumination, allowing the empty fog-muted harbor around their lateral separation to carry the isolation and lack of escape.

PRIORITY OF INSTRUCTIONS: the VISUAL APPROACH above governs camera,
framing and composition ONLY. Every other instruction in this prompt —
location, time of day, people and their identity, poses, props,
carried state, physical rules, and the no-text rule — is binding
exactly as written earlier. If the VISUAL APPROACH appears to conflict
with any of them, the earlier instruction wins.

Edit this photograph. Apply ONLY the corrections listed below,
each in place, changing nothing else:

CORRECTIONS:
- Change In-woo's pose so he is actively holding and securing the mooring line with his hands.
- Remove the distant boat visible in the fog on the left side to make the harbor completely empty.
- Replace the concrete pier ground with the dark, wet rocky surface shown in the previous shot still.

PRESERVE EVERYTHING ELSE EXACTLY as in the original photograph —
every person, pose, face, garment, object, set feature, lighting,
camera and framing stay identical. No text, captions or
watermarks anywhere.

Create ONE FINAL photorealistic live-action film still of the moment below — contemporary South Korea, 2026; all people are Korean unless stated. TIME OF DAY (lock): dusk, foggy.

SHOT TEXT (authoritative, Korean): 수리영이 절박한 눈빛으로 인우를 똑바로 올려다보는 얼굴 클로즈업

LOCATION (lock): At the fogbound dock beside the boat's mooring point, where she pleads with the boatman at dusk. The shot takes place here — the attached PREVIOUS SHOT STILL shows this exact place.

CAMERA & FRAME (follow exactly — this is the composition authority for this still; the reference images supply identity and place, never the framing):
- CAMERA: Just below 수리영’s eye line on her forward side, the dolly-in finishes on a close facial frame, placing her eyes near the upper center as they lift toward 인우. A restrained portion of 인우’s shoulder and jaw remains soft at the upper-right edge as an eyeline anchor, while 수리영’s face occupies most of the frame without becoming frontal.
- FRAMING SCALE: close-up
- FRAME LAYOUT: 수리영 in the middle-center of the frame, foreground, looks toward 인우; 인우 eyeline anchor in the upper-right of the frame, foreground.
- KEY BACKGROUND ELEMENTS: Fog-bound harbor (Distant detail obscured by fog); used as Soft, minimal background context behind 수리영's face; Fishing boat (Moored) — Only an indistinct dockside portion remains behind the characters; used as Peripheral spatial context at the edge of the close frame.
Compose the frame exactly as specified above — camera angle, subject scale and screen placement. Keep true physical scale between people and background elements: fixtures and distant objects occupy only the small screen area their real size and distance dictate; never enlarge a background object into a foreground presence.

THIS SHOT CONTINUES THE PREVIOUS SHOT: everything the attached still established — the place, its fixed features and wear, each person's clothing and state — persists. Any person in it who cannot move stays PRECISELY as photographed (body, pose, contact points, held objects); only the camera changes.

PREVIOUS STILL USAGE (follow exactly — what to take from the attached still and what to exclude): Take the same foggy dock, moored-boat surroundings, damp dusk atmosphere, and the two people’s continuing clothing and relative positions. Exclude the man’s back as a dominant foreground element and omit rope-fastening action; frame the young woman looking up at him desperately.

REALIZE FIGURATIVE LANGUAGE AS A LIVE-ACTION SHOT: the Korean shot
text may describe characters metaphorically, figuratively or with
exaggeration. Photograph what a real movie camera would actually
record on a physical set — exaggerated or figurative impressions
become realistic staging choices (lighting, distance, angle,
wardrobe), not literal fantasy imagery.
EVERY CHARACTER IS A HUMAN BEING: unless the story explicitly
features non-human or virtual beings (as in science fiction or
fantasy), every character — however indirectly, vaguely or
figuratively the text describes them — IS a real human. When the
text gives no direct visual description of a person, IMAGINE one
and still show them as a concrete, fully-formed human being:
always render the human form to the maximum extent — build,
posture, face, hands, clothing — never reduce a person to a shape,
blob, solid silhouette or abstract mass.

EXPRESSIONS ARE ACTED, NEVER ANATOMICAL: when the text describes a
person's eyes, face or presence emotionally or figuratively (vacant,
hollow, dazed, burning, lifeless gaze and the like), it describes an
ACTOR'S PERFORMANCE captured by a real camera — realize it ONLY
through gaze direction, focus, eyelids, facial muscles, stillness
and posture. Every living person's eyes remain anatomically normal
human eyes with a natural iris and pupil, natural sclera and normal
proportions; NEVER whiten, blank out, cloud over, glow, enlarge or
otherwise alter eyeballs, skin or anatomy — unless the story
explicitly declares that being non-human or supernatural in form.

PROPS FACE THE RIGHT WAY: every handheld or used object must be
oriented exactly as its real-world use requires. A person reading,
watching or operating something (a phone, a photograph, a paper,
any device) has its functional side — screen, front, page — facing
THEIR OWN eyes; the camera then sees whatever side the staging
geometry implies (often its back). Show the functional side to the
camera ONLY when the shot text itself stages it toward the viewer.
Never flip, mirror or reverse an object's front and back.

CARRIED STATE (persist exactly — must match the neighbouring shots of this scene): Su-ri-young still carries the old photograph, the backpack holding her printed maps, the coordinate note, and her phone as she pleads for passage.

PEOPLE: the SHOT TEXT alone decides whether any person is visible in this shot. IF a person appears, they must be one of: 수리영 (한국인 20대 초반 여성, 긴 검은 생머리, 젊은 타원형 얼굴) — never anyone else, and never add a person the shot text does not show. IF only part of a person is in frame (a hand, arm, foot, back, silhouette), that body part belongs to the specific person the shot text names — its sex, age, build, skin and grooming must unmistakably match that person's profile above.

No text, captions, watermarks or annotations anywhere.

Create ONE FINAL photorealistic live-action film still of the moment below — contemporary South Korea, 2026; all people are Korean unless stated. TIME OF DAY (lock): dusk, foggy.

SHOT TEXT (authoritative, Korean): 수리영이 절박한 눈빛으로 인우를 똑바로 올려다보는 얼굴 클로즈업

LOCATION (lock): At the fogbound dock beside the boat's mooring point, where she pleads with the boatman at dusk. The shot takes place here — the attached PREVIOUS SHOT STILL shows this exact place.

CAMERA & FRAME (follow exactly — this is the composition authority for this still; the reference images supply identity and place, never the framing):
- CAMERA: Just below 수리영’s eye line on her forward side, the dolly-in finishes on a close facial frame, placing her eyes near the upper center as they lift toward 인우. A restrained portion of 인우’s shoulder and jaw remains soft at the upper-right edge as an eyeline anchor, while 수리영’s face occupies most of the frame without becoming frontal.
- FRAMING SCALE: close-up
- FRAME LAYOUT: 수리영 in the middle-center of the frame, foreground, looks toward 인우; 인우 eyeline anchor in the upper-right of the frame, foreground.
- KEY BACKGROUND ELEMENTS: Fog-bound harbor (Distant detail obscured by fog); used as Soft, minimal background context behind 수리영's face; Fishing boat (Moored) — Only an indistinct dockside portion remains behind the characters; used as Peripheral spatial context at the edge of the close frame.
Compose the frame exactly as specified above — camera angle, subject scale and screen placement. Keep true physical scale between people and background elements: fixtures and distant objects occupy only the small screen area their real size and distance dictate; never enlarge a background object into a foreground presence.

THIS SHOT CONTINUES THE PREVIOUS SHOT: everything the attached still established — the place, its fixed features and wear, each person's clothing and state — persists. Any person in it who cannot move stays PRECISELY as photographed (body, pose, contact points, held objects); only the camera changes.

PREVIOUS STILL USAGE (follow exactly — what to take from the attached still and what to exclude): Take the same foggy dock, moored-boat surroundings, damp dusk atmosphere, and the two people’s continuing clothing and relative positions. Exclude the man’s back as a dominant foreground element and omit rope-fastening action; frame the young woman looking up at him desperately.

REALIZE FIGURATIVE LANGUAGE AS A LIVE-ACTION SHOT: the Korean shot
text may describe characters metaphorically, figuratively or with
exaggeration. Photograph what a real movie camera would actually
record on a physical set — exaggerated or figurative impressions
become realistic staging choices (lighting, distance, angle,
wardrobe), not literal fantasy imagery.
EVERY CHARACTER IS A HUMAN BEING: unless the story explicitly
features non-human or virtual beings (as in science fiction or
fantasy), every character — however indirectly, vaguely or
figuratively the text describes them — IS a real human. When the
text gives no direct visual description of a person, IMAGINE one
and still show them as a concrete, fully-formed human being:
always render the human form to the maximum extent — build,
posture, face, hands, clothing — never reduce a person to a shape,
blob, solid silhouette or abstract mass.

EXPRESSIONS ARE ACTED, NEVER ANATOMICAL: when the text describes a
person's eyes, face or presence emotionally or figuratively (vacant,
hollow, dazed, burning, lifeless gaze and the like), it describes an
ACTOR'S PERFORMANCE captured by a real camera — realize it ONLY
through gaze direction, focus, eyelids, facial muscles, stillness
and posture. Every living person's eyes remain anatomically normal
human eyes with a natural iris and pupil, natural sclera and normal
proportions; NEVER whiten, blank out, cloud over, glow, enlarge or
otherwise alter eyeballs, skin or anatomy — unless the story
explicitly declares that being non-human or supernatural in form.

PROPS FACE THE RIGHT WAY: every handheld or used object must be
oriented exactly as its real-world use requires. A person reading,
watching or operating something (a phone, a photograph, a paper,
any device) has its functional side — screen, front, page — facing
THEIR OWN eyes; the camera then sees whatever side the staging
geometry implies (often its back). Show the functional side to the
camera ONLY when the shot text itself stages it toward the viewer.
Never flip, mirror or reverse an object's front and back.

CARRIED STATE (persist exactly — must match the neighbouring shots of this scene): Su-ri-young still carries the old photograph, the backpack holding her printed maps, the coordinate note, and her phone as she pleads for passage.

PEOPLE: the SHOT TEXT alone decides whether any person is visible in this shot. IF a person appears, they must be one of: 수리영 (한국인 20대 초반 여성, 긴 검은 생머리, 젊은 타원형 얼굴) — never anyone else, and never add a person the shot text does not show. IF only part of a person is in frame (a hand, arm, foot, back, silhouette), that body part belongs to the specific person the shot text names — its sex, age, build, skin and grooming must unmistakably match that person's profile above.

No text, captions, watermarks or annotations anywhere.

VISUAL APPROACH — this take (camera, framing, composition):
Use an intimate, naturalistic close-focus lens feel with extremely selective depth, holding crisp focus on Su-ri-young’s lifted eyes while her features fall off gently toward the edges. Let In-woo’s restrained shoulder-and-jaw anchor dissolve softly at the upper right, with the fogbound harbor and moored boat reduced to muted, damp dusk shapes. Soft diffuse dusk light wraps her face, preserving delicate facial tension and a small natural catchlight without glamour.

PRIORITY OF INSTRUCTIONS: the VISUAL APPROACH above governs camera,
framing and composition ONLY. Every other instruction in this prompt —
location, time of day, people and their identity, poses, props,
carried state, physical rules, and the no-text rule — is binding
exactly as written earlier. If the VISUAL APPROACH appears to conflict
with any of them, the earlier instruction wins.

Create ONE FINAL photorealistic live-action film still of the moment below — contemporary South Korea, 2026; all people are Korean unless stated. TIME OF DAY (lock): dusk, foggy.

SHOT TEXT (authoritative, Korean): 수리영이 절박한 눈빛으로 인우를 똑바로 올려다보는 얼굴 클로즈업

LOCATION (lock): At the fogbound dock beside the boat's mooring point, where she pleads with the boatman at dusk. The shot takes place here — the attached PREVIOUS SHOT STILL shows this exact place.

CAMERA & FRAME (follow exactly — this is the composition authority for this still; the reference images supply identity and place, never the framing):
- CAMERA: Just below 수리영’s eye line on her forward side, the dolly-in finishes on a close facial frame, placing her eyes near the upper center as they lift toward 인우. A restrained portion of 인우’s shoulder and jaw remains soft at the upper-right edge as an eyeline anchor, while 수리영’s face occupies most of the frame without becoming frontal.
- FRAMING SCALE: close-up
- FRAME LAYOUT: 수리영 in the middle-center of the frame, foreground, looks toward 인우; 인우 eyeline anchor in the upper-right of the frame, foreground.
- KEY BACKGROUND ELEMENTS: Fog-bound harbor (Distant detail obscured by fog); used as Soft, minimal background context behind 수리영's face; Fishing boat (Moored) — Only an indistinct dockside portion remains behind the characters; used as Peripheral spatial context at the edge of the close frame.
Compose the frame exactly as specified above — camera angle, subject scale and screen placement. Keep true physical scale between people and background elements: fixtures and distant objects occupy only the small screen area their real size and distance dictate; never enlarge a background object into a foreground presence.

THIS SHOT CONTINUES THE PREVIOUS SHOT: everything the attached still established — the place, its fixed features and wear, each person's clothing and state — persists. Any person in it who cannot move stays PRECISELY as photographed (body, pose, contact points, held objects); only the camera changes.

PREVIOUS STILL USAGE (follow exactly — what to take from the attached still and what to exclude): Take the same foggy dock, moored-boat surroundings, damp dusk atmosphere, and the two people’s continuing clothing and relative positions. Exclude the man’s back as a dominant foreground element and omit rope-fastening action; frame the young woman looking up at him desperately.

REALIZE FIGURATIVE LANGUAGE AS A LIVE-ACTION SHOT: the Korean shot
text may describe characters metaphorically, figuratively or with
exaggeration. Photograph what a real movie camera would actually
record on a physical set — exaggerated or figurative impressions
become realistic staging choices (lighting, distance, angle,
wardrobe), not literal fantasy imagery.
EVERY CHARACTER IS A HUMAN BEING: unless the story explicitly
features non-human or virtual beings (as in science fiction or
fantasy), every character — however indirectly, vaguely or
figuratively the text describes them — IS a real human. When the
text gives no direct visual description of a person, IMAGINE one
and still show them as a concrete, fully-formed human being:
always render the human form to the maximum extent — build,
posture, face, hands, clothing — never reduce a person to a shape,
blob, solid silhouette or abstract mass.

EXPRESSIONS ARE ACTED, NEVER ANATOMICAL: when the text describes a
person's eyes, face or presence emotionally or figuratively (vacant,
hollow, dazed, burning, lifeless gaze and the like), it describes an
ACTOR'S PERFORMANCE captured by a real camera — realize it ONLY
through gaze direction, focus, eyelids, facial muscles, stillness
and posture. Every living person's eyes remain anatomically normal
human eyes with a natural iris and pupil, natural sclera and normal
proportions; NEVER whiten, blank out, cloud over, glow, enlarge or
otherwise alter eyeballs, skin or anatomy — unless the story
explicitly declares that being non-human or supernatural in form.

PROPS FACE THE RIGHT WAY: every handheld or used object must be
oriented exactly as its real-world use requires. A person reading,
watching or operating something (a phone, a photograph, a paper,
any device) has its functional side — screen, front, page — facing
THEIR OWN eyes; the camera then sees whatever side the staging
geometry implies (often its back). Show the functional side to the
camera ONLY when the shot text itself stages it toward the viewer.
Never flip, mirror or reverse an object's front and back.

CARRIED STATE (persist exactly — must match the neighbouring shots of this scene): Su-ri-young still carries the old photograph, the backpack holding her printed maps, the coordinate note, and her phone as she pleads for passage.

PEOPLE: the SHOT TEXT alone decides whether any person is visible in this shot. IF a person appears, they must be one of: 수리영 (한국인 20대 초반 여성, 긴 검은 생머리, 젊은 타원형 얼굴) — never anyone else, and never add a person the shot text does not show. IF only part of a person is in frame (a hand, arm, foot, back, silhouette), that body part belongs to the specific person the shot text names — its sex, age, build, skin and grooming must unmistakably match that person's profile above.

No text, captions, watermarks or annotations anywhere.

VISUAL APPROACH — this take (camera, framing, composition):
Give the close-up a subtly compressed lens feel, making the narrow space between Su-ri-young’s upturned face and In-woo’s soft upper-right anchor feel emotionally pressurized. Keep slightly more tonal definition across her whole face while maintaining shallow separation from the fog-obscured dock and indistinct boat edge. Shape the available dusk light from the side, leaving a restrained shadow gradient across her nonfrontal face so the fixed upward gaze and tightened facial muscles carry the desperation.

PRIORITY OF INSTRUCTIONS: the VISUAL APPROACH above governs camera,
framing and composition ONLY. Every other instruction in this prompt —
location, time of day, people and their identity, poses, props,
carried state, physical rules, and the no-text rule — is binding
exactly as written earlier. If the VISUAL APPROACH appears to conflict
with any of them, the earlier instruction wins.

Edit this photograph. Apply ONLY the corrections listed below,
each in place, changing nothing else:

CORRECTIONS:
- Add the dark grey beanie from the previous shot to the man's head on the right edge of the frame.

PRESERVE EVERYTHING ELSE EXACTLY as in the original photograph —
every person, pose, face, garment, object, set feature, lighting,
camera and framing stay identical. No text, captions or
watermarks anywhere.

Create ONE FINAL photorealistic live-action film still of the moment below — contemporary South Korea, 2026; all people are Korean unless stated. TIME OF DAY (lock): dusk.

SHOT TEXT (authoritative, Korean): 혜수가 엉망이 된 옥탑방 거실에 서서 휴대폰을 귀에 댄 채 굳어 있는 상체

LOCATION (lock): Inside the ransacked rooftop apartment living room at dusk, where the detective listens to the recorded message. The shot takes place here — the attached PREVIOUS SHOT STILL shows this exact place.

CAMERA & FRAME (follow exactly — this is the composition authority for this still; the reference images supply identity and place, never the framing):
- CAMERA: At 혜수’s upper-body height, the dolly-in settles at a medium distance directly before her, framing her from the upper torso upward with the phone retained against her ear. She holds a three-quarter turn toward 김형사 off screen, while fragments of the ransacked room remain around her shoulders rather than competing with her frozen expression.
- FRAMING SCALE: medium shot
- KEY BACKGROUND ELEMENTS: Mobile phone (Held to her ear while she checks the recorded message) — Its earpiece side is pressed to 혜수's ear while part of the outer face remains visible to camera; used as Essential action prop held beside 혜수's face; Scattered household objects (Scattered across the room) — Displaced sides and edges remain visible behind 혜수; used as Partial background evidence of the earlier search.
Compose the frame exactly as specified above — camera angle, subject scale and screen placement. Keep true physical scale between people and background elements: fixtures and distant objects occupy only the small screen area their real size and distance dictate; never enlarge a background object into a foreground presence.

THIS SHOT CONTINUES THE PREVIOUS SHOT: everything the attached still established — the place, its fixed features and wear, each person's clothing and state — persists. Any person in it who cannot move stays PRECISELY as photographed (body, pose, contact points, held objects); only the camera changes.

PREVIOUS STILL USAGE (follow exactly — what to take from the attached still and what to exclude): Take the ransacked rooftop-room living area, open drawers, overturned furniture, open window, dusk lighting, and persistent disorder. Exclude the younger woman and her searching action; include only the female officer standing in the room with a phone to her ear.

REALIZE FIGURATIVE LANGUAGE AS A LIVE-ACTION SHOT: the Korean shot
text may describe characters metaphorically, figuratively or with
exaggeration. Photograph what a real movie camera would actually
record on a physical set — exaggerated or figurative impressions
become realistic staging choices (lighting, distance, angle,
wardrobe), not literal fantasy imagery.
EVERY CHARACTER IS A HUMAN BEING: unless the story explicitly
features non-human or virtual beings (as in science fiction or
fantasy), every character — however indirectly, vaguely or
figuratively the text describes them — IS a real human. When the
text gives no direct visual description of a person, IMAGINE one
and still show them as a concrete, fully-formed human being:
always render the human form to the maximum extent — build,
posture, face, hands, clothing — never reduce a person to a shape,
blob, solid silhouette or abstract mass.

EXPRESSIONS ARE ACTED, NEVER ANATOMICAL: when the text describes a
person's eyes, face or presence emotionally or figuratively (vacant,
hollow, dazed, burning, lifeless gaze and the like), it describes an
ACTOR'S PERFORMANCE captured by a real camera — realize it ONLY
through gaze direction, focus, eyelids, facial muscles, stillness
and posture. Every living person's eyes remain anatomically normal
human eyes with a natural iris and pupil, natural sclera and normal
proportions; NEVER whiten, blank out, cloud over, glow, enlarge or
otherwise alter eyeballs, skin or anatomy — unless the story
explicitly declares that being non-human or supernatural in form.

PROPS FACE THE RIGHT WAY: every handheld or used object must be
oriented exactly as its real-world use requires. A person reading,
watching or operating something (a phone, a photograph, a paper,
any device) has its functional side — screen, front, page — facing
THEIR OWN eyes; the camera then sees whatever side the staging
geometry implies (often its back). Show the functional side to the
camera ONLY when the shot text itself stages it toward the viewer.
Never flip, mirror or reverse an object's front and back.

CARRIED STATE (persist exactly — must match the neighbouring shots of this scene): The rooftop home remains ransacked, with footprints, displaced furnishings, an open window, and overturned chairs; the red circle remains on the bedroom wall, but Su-ri-young has taken the old photograph. Hye-su holds her phone while listening to Su-ri-young’s recorded message.

PEOPLE: the SHOT TEXT alone decides whether any person is visible in this shot. IF a person appears, they must be one of: 혜수 (한국인 40대 초반 여성, 단정한 검은 머리, 성숙한 타원형 얼굴) — never anyone else, and never add a person the shot text does not show. IF only part of a person is in frame (a hand, arm, foot, back, silhouette), that body part belongs to the specific person the shot text names — its sex, age, build, skin and grooming must unmistakably match that person's profile above.

No text, captions, watermarks or annotations anywhere.

Create ONE FINAL photorealistic live-action film still of the moment below — contemporary South Korea, 2026; all people are Korean unless stated. TIME OF DAY (lock): dusk.

SHOT TEXT (authoritative, Korean): 혜수가 엉망이 된 옥탑방 거실에 서서 휴대폰을 귀에 댄 채 굳어 있는 상체

LOCATION (lock): Inside the ransacked rooftop apartment living room at dusk, where the detective listens to the recorded message. The shot takes place here — the attached PREVIOUS SHOT STILL shows this exact place.

CAMERA & FRAME (follow exactly — this is the composition authority for this still; the reference images supply identity and place, never the framing):
- CAMERA: At 혜수’s upper-body height, the dolly-in settles at a medium distance directly before her, framing her from the upper torso upward with the phone retained against her ear. She holds a three-quarter turn toward 김형사 off screen, while fragments of the ransacked room remain around her shoulders rather than competing with her frozen expression.
- FRAMING SCALE: medium shot
- KEY BACKGROUND ELEMENTS: Mobile phone (Held to her ear while she checks the recorded message) — Its earpiece side is pressed to 혜수's ear while part of the outer face remains visible to camera; used as Essential action prop held beside 혜수's face; Scattered household objects (Scattered across the room) — Displaced sides and edges remain visible behind 혜수; used as Partial background evidence of the earlier search.
Compose the frame exactly as specified above — camera angle, subject scale and screen placement. Keep true physical scale between people and background elements: fixtures and distant objects occupy only the small screen area their real size and distance dictate; never enlarge a background object into a foreground presence.

THIS SHOT CONTINUES THE PREVIOUS SHOT: everything the attached still established — the place, its fixed features and wear, each person's clothing and state — persists. Any person in it who cannot move stays PRECISELY as photographed (body, pose, contact points, held objects); only the camera changes.

PREVIOUS STILL USAGE (follow exactly — what to take from the attached still and what to exclude): Take the ransacked rooftop-room living area, open drawers, overturned furniture, open window, dusk lighting, and persistent disorder. Exclude the younger woman and her searching action; include only the female officer standing in the room with a phone to her ear.

REALIZE FIGURATIVE LANGUAGE AS A LIVE-ACTION SHOT: the Korean shot
text may describe characters metaphorically, figuratively or with
exaggeration. Photograph what a real movie camera would actually
record on a physical set — exaggerated or figurative impressions
become realistic staging choices (lighting, distance, angle,
wardrobe), not literal fantasy imagery.
EVERY CHARACTER IS A HUMAN BEING: unless the story explicitly
features non-human or virtual beings (as in science fiction or
fantasy), every character — however indirectly, vaguely or
figuratively the text describes them — IS a real human. When the
text gives no direct visual description of a person, IMAGINE one
and still show them as a concrete, fully-formed human being:
always render the human form to the maximum extent — build,
posture, face, hands, clothing — never reduce a person to a shape,
blob, solid silhouette or abstract mass.

EXPRESSIONS ARE ACTED, NEVER ANATOMICAL: when the text describes a
person's eyes, face or presence emotionally or figuratively (vacant,
hollow, dazed, burning, lifeless gaze and the like), it describes an
ACTOR'S PERFORMANCE captured by a real camera — realize it ONLY
through gaze direction, focus, eyelids, facial muscles, stillness
and posture. Every living person's eyes remain anatomically normal
human eyes with a natural iris and pupil, natural sclera and normal
proportions; NEVER whiten, blank out, cloud over, glow, enlarge or
otherwise alter eyeballs, skin or anatomy — unless the story
explicitly declares that being non-human or supernatural in form.

PROPS FACE THE RIGHT WAY: every handheld or used object must be
oriented exactly as its real-world use requires. A person reading,
watching or operating something (a phone, a photograph, a paper,
any device) has its functional side — screen, front, page — facing
THEIR OWN eyes; the camera then sees whatever side the staging
geometry implies (often its back). Show the functional side to the
camera ONLY when the shot text itself stages it toward the viewer.
Never flip, mirror or reverse an object's front and back.

CARRIED STATE (persist exactly — must match the neighbouring shots of this scene): The rooftop home remains ransacked, with footprints, displaced furnishings, an open window, and overturned chairs; the red circle remains on the bedroom wall, but Su-ri-young has taken the old photograph. Hye-su holds her phone while listening to Su-ri-young’s recorded message.

PEOPLE: the SHOT TEXT alone decides whether any person is visible in this shot. IF a person appears, they must be one of: 혜수 (한국인 40대 초반 여성, 단정한 검은 머리, 성숙한 타원형 얼굴) — never anyone else, and never add a person the shot text does not show. IF only part of a person is in frame (a hand, arm, foot, back, silhouette), that body part belongs to the specific person the shot text names — its sex, age, build, skin and grooming must unmistakably match that person's profile above.

No text, captions, watermarks or annotations anywhere.

VISUAL APPROACH — this take (camera, framing, composition):
Use a gently compressed, intimate medium-shot perspective with very shallow depth of field, holding critical focus on Hye-su’s near eye, restrained facial tension, and the phone beside her face. Let the existing dusk from the open window wrap softly across her features with a faint edge along her hair, while the surrounding disorder falls into muted, recognizable fragments. Keep the settled frame unnervingly still, making her arrested posture carry the beat.

PRIORITY OF INSTRUCTIONS: the VISUAL APPROACH above governs camera,
framing and composition ONLY. Every other instruction in this prompt —
location, time of day, people and their identity, poses, props,
carried state, physical rules, and the no-text rule — is binding
exactly as written earlier. If the VISUAL APPROACH appears to conflict
with any of them, the earlier instruction wins.

Create ONE FINAL photorealistic live-action film still of the moment below — contemporary South Korea, 2026; all people are Korean unless stated. TIME OF DAY (lock): dusk.

SHOT TEXT (authoritative, Korean): 혜수가 엉망이 된 옥탑방 거실에 서서 휴대폰을 귀에 댄 채 굳어 있는 상체

LOCATION (lock): Inside the ransacked rooftop apartment living room at dusk, where the detective listens to the recorded message. The shot takes place here — the attached PREVIOUS SHOT STILL shows this exact place.

CAMERA & FRAME (follow exactly — this is the composition authority for this still; the reference images supply identity and place, never the framing):
- CAMERA: At 혜수’s upper-body height, the dolly-in settles at a medium distance directly before her, framing her from the upper torso upward with the phone retained against her ear. She holds a three-quarter turn toward 김형사 off screen, while fragments of the ransacked room remain around her shoulders rather than competing with her frozen expression.
- FRAMING SCALE: medium shot
- KEY BACKGROUND ELEMENTS: Mobile phone (Held to her ear while she checks the recorded message) — Its earpiece side is pressed to 혜수's ear while part of the outer face remains visible to camera; used as Essential action prop held beside 혜수's face; Scattered household objects (Scattered across the room) — Displaced sides and edges remain visible behind 혜수; used as Partial background evidence of the earlier search.
Compose the frame exactly as specified above — camera angle, subject scale and screen placement. Keep true physical scale between people and background elements: fixtures and distant objects occupy only the small screen area their real size and distance dictate; never enlarge a background object into a foreground presence.

THIS SHOT CONTINUES THE PREVIOUS SHOT: everything the attached still established — the place, its fixed features and wear, each person's clothing and state — persists. Any person in it who cannot move stays PRECISELY as photographed (body, pose, contact points, held objects); only the camera changes.

PREVIOUS STILL USAGE (follow exactly — what to take from the attached still and what to exclude): Take the ransacked rooftop-room living area, open drawers, overturned furniture, open window, dusk lighting, and persistent disorder. Exclude the younger woman and her searching action; include only the female officer standing in the room with a phone to her ear.

REALIZE FIGURATIVE LANGUAGE AS A LIVE-ACTION SHOT: the Korean shot
text may describe characters metaphorically, figuratively or with
exaggeration. Photograph what a real movie camera would actually
record on a physical set — exaggerated or figurative impressions
become realistic staging choices (lighting, distance, angle,
wardrobe), not literal fantasy imagery.
EVERY CHARACTER IS A HUMAN BEING: unless the story explicitly
features non-human or virtual beings (as in science fiction or
fantasy), every character — however indirectly, vaguely or
figuratively the text describes them — IS a real human. When the
text gives no direct visual description of a person, IMAGINE one
and still show them as a concrete, fully-formed human being:
always render the human form to the maximum extent — build,
posture, face, hands, clothing — never reduce a person to a shape,
blob, solid silhouette or abstract mass.

EXPRESSIONS ARE ACTED, NEVER ANATOMICAL: when the text describes a
person's eyes, face or presence emotionally or figuratively (vacant,
hollow, dazed, burning, lifeless gaze and the like), it describes an
ACTOR'S PERFORMANCE captured by a real camera — realize it ONLY
through gaze direction, focus, eyelids, facial muscles, stillness
and posture. Every living person's eyes remain anatomically normal
human eyes with a natural iris and pupil, natural sclera and normal
proportions; NEVER whiten, blank out, cloud over, glow, enlarge or
otherwise alter eyeballs, skin or anatomy — unless the story
explicitly declares that being non-human or supernatural in form.

PROPS FACE THE RIGHT WAY: every handheld or used object must be
oriented exactly as its real-world use requires. A person reading,
watching or operating something (a phone, a photograph, a paper,
any device) has its functional side — screen, front, page — facing
THEIR OWN eyes; the camera then sees whatever side the staging
geometry implies (often its back). Show the functional side to the
camera ONLY when the shot text itself stages it toward the viewer.
Never flip, mirror or reverse an object's front and back.

CARRIED STATE (persist exactly — must match the neighbouring shots of this scene): The rooftop home remains ransacked, with footprints, displaced furnishings, an open window, and overturned chairs; the red circle remains on the bedroom wall, but Su-ri-young has taken the old photograph. Hye-su holds her phone while listening to Su-ri-young’s recorded message.

PEOPLE: the SHOT TEXT alone decides whether any person is visible in this shot. IF a person appears, they must be one of: 혜수 (한국인 40대 초반 여성, 단정한 검은 머리, 성숙한 타원형 얼굴) — never anyone else, and never add a person the shot text does not show. IF only part of a person is in frame (a hand, arm, foot, back, silhouette), that body part belongs to the specific person the shot text names — its sex, age, build, skin and grooming must unmistakably match that person's profile above.

No text, captions, watermarks or annotations anywhere.

VISUAL APPROACH — this take (camera, framing, composition):
Photograph the medium shot with a natural perspective and broader depth, arranging the room’s displaced edges as subdued geometric layers around Hye-su’s shoulders without drawing attention from her face. Preserve the dim dusk ambience with restrained contrast, allowing the open-window light to shape the space while her expression remains clearly readable. Favor a firm, observational composition in which her rigid stillness stands against the visible disorder behind her.

PRIORITY OF INSTRUCTIONS: the VISUAL APPROACH above governs camera,
framing and composition ONLY. Every other instruction in this prompt —
location, time of day, people and their identity, poses, props,
carried state, physical rules, and the no-text rule — is binding
exactly as written earlier. If the VISUAL APPROACH appears to conflict
with any of them, the earlier instruction wins.

Edit this photograph. Apply ONLY the corrections listed below,
each in place, changing nothing else:

CORRECTIONS:
- Rotate the character's torso to hold a three-quarter turn away from the camera.
- Change the background window to an open state matching the previous shot reference.

PRESERVE EVERYTHING ELSE EXACTLY as in the original photograph —
every person, pose, face, garment, object, set feature, lighting,
camera and framing stay identical. No text, captions or
watermarks anywhere.

Create ONE FINAL photorealistic live-action film still of the moment below — contemporary South Korea, 2026; all people are Korean unless stated. TIME OF DAY (lock): dusk.

SHOT TEXT (authoritative, Korean): 물건들이 나뒹구는 텅 빈 옥탑방 거실에 혜수와 김형사가 우두커니 서 있는 넓은 구도

LOCATION (lock): Inside the desolate, ransacked living room of the rooftop apartment, facing the wind-rattled entrance door at dusk. The shot takes place here — the attached PREVIOUS SHOT STILL shows this exact place.

CAMERA & FRAME (follow exactly — this is the composition authority for this still; the reference images supply identity and place, never the framing):
- CAMERA: At eye level near the entrance, the dolly-out reaches its final wide distance and holds, leaving 혜수 and 김형사 separated within the middle of the otherwise empty living room. Scattered objects occupy the lower frame and the wind-rattled entrance door remains at the edge, allowing vacant space to outweigh both figures.
- FRAMING SCALE: wide shot
- KEY BACKGROUND ELEMENTS: Scattered household objects (Scattered across the room) — Their displaced faces and edges point in unrelated directions around the two figures; used as Lower-frame evidence of the room's disruption and surrounding negative space; Entrance door (Rattling in the wind) — Its interior-facing side is visible near the room entrance; used as Edge-of-frame source of the repeated rattling sound; Empty living room (Ransacked and largely vacant); used as Broad spatial field isolating the two motionless figures.
Compose the frame exactly as specified above — camera angle, subject scale and screen placement. Keep true physical scale between people and background elements: fixtures and distant objects occupy only the small screen area their real size and distance dictate; never enlarge a background object into a foreground presence.

THIS SHOT CONTINUES THE PREVIOUS SHOT: everything the attached still established — the place, its fixed features and wear, each person's clothing and state — persists. Any person in it who cannot move stays PRECISELY as photographed (body, pose, contact points, held objects); only the camera changes.

PREVIOUS STILL USAGE (follow exactly — what to take from the attached still and what to exclude): Take the same ransacked living-room layout, scattered household items, overturned chairs, open window, and desolate dusk light. Exclude the younger woman, backpack, photograph, and all earlier searching action; place only the two investigators standing still in the empty room.

REALIZE FIGURATIVE LANGUAGE AS A LIVE-ACTION SHOT: the Korean shot
text may describe characters metaphorically, figuratively or with
exaggeration. Photograph what a real movie camera would actually
record on a physical set — exaggerated or figurative impressions
become realistic staging choices (lighting, distance, angle,
wardrobe), not literal fantasy imagery.
EVERY CHARACTER IS A HUMAN BEING: unless the story explicitly
features non-human or virtual beings (as in science fiction or
fantasy), every character — however indirectly, vaguely or
figuratively the text describes them — IS a real human. When the
text gives no direct visual description of a person, IMAGINE one
and still show them as a concrete, fully-formed human being:
always render the human form to the maximum extent — build,
posture, face, hands, clothing — never reduce a person to a shape,
blob, solid silhouette or abstract mass.

EXPRESSIONS ARE ACTED, NEVER ANATOMICAL: when the text describes a
person's eyes, face or presence emotionally or figuratively (vacant,
hollow, dazed, burning, lifeless gaze and the like), it describes an
ACTOR'S PERFORMANCE captured by a real camera — realize it ONLY
through gaze direction, focus, eyelids, facial muscles, stillness
and posture. Every living person's eyes remain anatomically normal
human eyes with a natural iris and pupil, natural sclera and normal
proportions; NEVER whiten, blank out, cloud over, glow, enlarge or
otherwise alter eyeballs, skin or anatomy — unless the story
explicitly declares that being non-human or supernatural in form.

PROPS FACE THE RIGHT WAY: every handheld or used object must be
oriented exactly as its real-world use requires. A person reading,
watching or operating something (a phone, a photograph, a paper,
any device) has its functional side — screen, front, page — facing
THEIR OWN eyes; the camera then sees whatever side the staging
geometry implies (often its back). Show the functional side to the
camera ONLY when the shot text itself stages it toward the viewer.
Never flip, mirror or reverse an object's front and back.

CARRIED STATE (persist exactly — must match the neighbouring shots of this scene): The empty rooftop home remains ransacked, with its furniture and household objects displaced; the old photograph is no longer there because Su-ri-young took it. The red circle remains on the bedroom wall.

PEOPLE: the SHOT TEXT alone decides whether any person is visible in this shot. IF a person appears, they must be one of: 혜수 (한국인 40대 초반 여성, 단정한 검은 머리, 성숙한 타원형 얼굴); 김형사 (한국인 30대 남성, 짧은 검은 머리, 단정한 머리 모양, 성숙한 얼굴) — never anyone else, and never add a person the shot text does not show. IF only part of a person is in frame (a hand, arm, foot, back, silhouette), that body part belongs to the specific person the shot text names — its sex, age, build, skin and grooming must unmistakably match that person's profile above.

No text, captions, watermarks or annotations anywhere.

Create ONE FINAL photorealistic live-action film still of the moment below — contemporary South Korea, 2026; all people are Korean unless stated. TIME OF DAY (lock): dusk.

SHOT TEXT (authoritative, Korean): 물건들이 나뒹구는 텅 빈 옥탑방 거실에 혜수와 김형사가 우두커니 서 있는 넓은 구도

LOCATION (lock): Inside the desolate, ransacked living room of the rooftop apartment, facing the wind-rattled entrance door at dusk. The shot takes place here — the attached PREVIOUS SHOT STILL shows this exact place.

CAMERA & FRAME (follow exactly — this is the composition authority for this still; the reference images supply identity and place, never the framing):
- CAMERA: At eye level near the entrance, the dolly-out reaches its final wide distance and holds, leaving 혜수 and 김형사 separated within the middle of the otherwise empty living room. Scattered objects occupy the lower frame and the wind-rattled entrance door remains at the edge, allowing vacant space to outweigh both figures.
- FRAMING SCALE: wide shot
- KEY BACKGROUND ELEMENTS: Scattered household objects (Scattered across the room) — Their displaced faces and edges point in unrelated directions around the two figures; used as Lower-frame evidence of the room's disruption and surrounding negative space; Entrance door (Rattling in the wind) — Its interior-facing side is visible near the room entrance; used as Edge-of-frame source of the repeated rattling sound; Empty living room (Ransacked and largely vacant); used as Broad spatial field isolating the two motionless figures.
Compose the frame exactly as specified above — camera angle, subject scale and screen placement. Keep true physical scale between people and background elements: fixtures and distant objects occupy only the small screen area their real size and distance dictate; never enlarge a background object into a foreground presence.

THIS SHOT CONTINUES THE PREVIOUS SHOT: everything the attached still established — the place, its fixed features and wear, each person's clothing and state — persists. Any person in it who cannot move stays PRECISELY as photographed (body, pose, contact points, held objects); only the camera changes.

PREVIOUS STILL USAGE (follow exactly — what to take from the attached still and what to exclude): Take the same ransacked living-room layout, scattered household items, overturned chairs, open window, and desolate dusk light. Exclude the younger woman, backpack, photograph, and all earlier searching action; place only the two investigators standing still in the empty room.

REALIZE FIGURATIVE LANGUAGE AS A LIVE-ACTION SHOT: the Korean shot
text may describe characters metaphorically, figuratively or with
exaggeration. Photograph what a real movie camera would actually
record on a physical set — exaggerated or figurative impressions
become realistic staging choices (lighting, distance, angle,
wardrobe), not literal fantasy imagery.
EVERY CHARACTER IS A HUMAN BEING: unless the story explicitly
features non-human or virtual beings (as in science fiction or
fantasy), every character — however indirectly, vaguely or
figuratively the text describes them — IS a real human. When the
text gives no direct visual description of a person, IMAGINE one
and still show them as a concrete, fully-formed human being:
always render the human form to the maximum extent — build,
posture, face, hands, clothing — never reduce a person to a shape,
blob, solid silhouette or abstract mass.

EXPRESSIONS ARE ACTED, NEVER ANATOMICAL: when the text describes a
person's eyes, face or presence emotionally or figuratively (vacant,
hollow, dazed, burning, lifeless gaze and the like), it describes an
ACTOR'S PERFORMANCE captured by a real camera — realize it ONLY
through gaze direction, focus, eyelids, facial muscles, stillness
and posture. Every living person's eyes remain anatomically normal
human eyes with a natural iris and pupil, natural sclera and normal
proportions; NEVER whiten, blank out, cloud over, glow, enlarge or
otherwise alter eyeballs, skin or anatomy — unless the story
explicitly declares that being non-human or supernatural in form.

PROPS FACE THE RIGHT WAY: every handheld or used object must be
oriented exactly as its real-world use requires. A person reading,
watching or operating something (a phone, a photograph, a paper,
any device) has its functional side — screen, front, page — facing
THEIR OWN eyes; the camera then sees whatever side the staging
geometry implies (often its back). Show the functional side to the
camera ONLY when the shot text itself stages it toward the viewer.
Never flip, mirror or reverse an object's front and back.

CARRIED STATE (persist exactly — must match the neighbouring shots of this scene): The empty rooftop home remains ransacked, with its furniture and household objects displaced; the old photograph is no longer there because Su-ri-young took it. The red circle remains on the bedroom wall.

PEOPLE: the SHOT TEXT alone decides whether any person is visible in this shot. IF a person appears, they must be one of: 혜수 (한국인 40대 초반 여성, 단정한 검은 머리, 성숙한 타원형 얼굴); 김형사 (한국인 30대 남성, 짧은 검은 머리, 단정한 머리 모양, 성숙한 얼굴) — never anyone else, and never add a person the shot text does not show. IF only part of a person is in frame (a hand, arm, foot, back, silhouette), that body part belongs to the specific person the shot text names — its sex, age, build, skin and grooming must unmistakably match that person's profile above.

No text, captions, watermarks or annotations anywhere.

VISUAL APPROACH — this take (camera, framing, composition):
Use a restrained, rectilinear wide-lens feel with deep focus and a level, nearly frontal composition. Hold 혜수 and 김형사 small and distinctly separated in the middle field, while the scattered objects form a sparse, disordered lower-frame pattern and the entrance door stays peripheral. Let the cool dusk illumination fall softly across the room with subdued contrast, making the broad vacancy feel heavier than either motionless figure.

PRIORITY OF INSTRUCTIONS: the VISUAL APPROACH above governs camera,
framing and composition ONLY. Every other instruction in this prompt —
location, time of day, people and their identity, poses, props,
carried state, physical rules, and the no-text rule — is binding
exactly as written earlier. If the VISUAL APPROACH appears to conflict
with any of them, the earlier instruction wins.

Create ONE FINAL photorealistic live-action film still of the moment below — contemporary South Korea, 2026; all people are Korean unless stated. TIME OF DAY (lock): dusk.

SHOT TEXT (authoritative, Korean): 물건들이 나뒹구는 텅 빈 옥탑방 거실에 혜수와 김형사가 우두커니 서 있는 넓은 구도

LOCATION (lock): Inside the desolate, ransacked living room of the rooftop apartment, facing the wind-rattled entrance door at dusk. The shot takes place here — the attached PREVIOUS SHOT STILL shows this exact place.

CAMERA & FRAME (follow exactly — this is the composition authority for this still; the reference images supply identity and place, never the framing):
- CAMERA: At eye level near the entrance, the dolly-out reaches its final wide distance and holds, leaving 혜수 and 김형사 separated within the middle of the otherwise empty living room. Scattered objects occupy the lower frame and the wind-rattled entrance door remains at the edge, allowing vacant space to outweigh both figures.
- FRAMING SCALE: wide shot
- KEY BACKGROUND ELEMENTS: Scattered household objects (Scattered across the room) — Their displaced faces and edges point in unrelated directions around the two figures; used as Lower-frame evidence of the room's disruption and surrounding negative space; Entrance door (Rattling in the wind) — Its interior-facing side is visible near the room entrance; used as Edge-of-frame source of the repeated rattling sound; Empty living room (Ransacked and largely vacant); used as Broad spatial field isolating the two motionless figures.
Compose the frame exactly as specified above — camera angle, subject scale and screen placement. Keep true physical scale between people and background elements: fixtures and distant objects occupy only the small screen area their real size and distance dictate; never enlarge a background object into a foreground presence.

THIS SHOT CONTINUES THE PREVIOUS SHOT: everything the attached still established — the place, its fixed features and wear, each person's clothing and state — persists. Any person in it who cannot move stays PRECISELY as photographed (body, pose, contact points, held objects); only the camera changes.

PREVIOUS STILL USAGE (follow exactly — what to take from the attached still and what to exclude): Take the same ransacked living-room layout, scattered household items, overturned chairs, open window, and desolate dusk light. Exclude the younger woman, backpack, photograph, and all earlier searching action; place only the two investigators standing still in the empty room.

REALIZE FIGURATIVE LANGUAGE AS A LIVE-ACTION SHOT: the Korean shot
text may describe characters metaphorically, figuratively or with
exaggeration. Photograph what a real movie camera would actually
record on a physical set — exaggerated or figurative impressions
become realistic staging choices (lighting, distance, angle,
wardrobe), not literal fantasy imagery.
EVERY CHARACTER IS A HUMAN BEING: unless the story explicitly
features non-human or virtual beings (as in science fiction or
fantasy), every character — however indirectly, vaguely or
figuratively the text describes them — IS a real human. When the
text gives no direct visual description of a person, IMAGINE one
and still show them as a concrete, fully-formed human being:
always render the human form to the maximum extent — build,
posture, face, hands, clothing — never reduce a person to a shape,
blob, solid silhouette or abstract mass.

EXPRESSIONS ARE ACTED, NEVER ANATOMICAL: when the text describes a
person's eyes, face or presence emotionally or figuratively (vacant,
hollow, dazed, burning, lifeless gaze and the like), it describes an
ACTOR'S PERFORMANCE captured by a real camera — realize it ONLY
through gaze direction, focus, eyelids, facial muscles, stillness
and posture. Every living person's eyes remain anatomically normal
human eyes with a natural iris and pupil, natural sclera and normal
proportions; NEVER whiten, blank out, cloud over, glow, enlarge or
otherwise alter eyeballs, skin or anatomy — unless the story
explicitly declares that being non-human or supernatural in form.

PROPS FACE THE RIGHT WAY: every handheld or used object must be
oriented exactly as its real-world use requires. A person reading,
watching or operating something (a phone, a photograph, a paper,
any device) has its functional side — screen, front, page — facing
THEIR OWN eyes; the camera then sees whatever side the staging
geometry implies (often its back). Show the functional side to the
camera ONLY when the shot text itself stages it toward the viewer.
Never flip, mirror or reverse an object's front and back.

CARRIED STATE (persist exactly — must match the neighbouring shots of this scene): The empty rooftop home remains ransacked, with its furniture and household objects displaced; the old photograph is no longer there because Su-ri-young took it. The red circle remains on the bedroom wall.

PEOPLE: the SHOT TEXT alone decides whether any person is visible in this shot. IF a person appears, they must be one of: 혜수 (한국인 40대 초반 여성, 단정한 검은 머리, 성숙한 타원형 얼굴); 김형사 (한국인 30대 남성, 짧은 검은 머리, 단정한 머리 모양, 성숙한 얼굴) — never anyone else, and never add a person the shot text does not show. IF only part of a person is in frame (a hand, arm, foot, back, silhouette), that body part belongs to the specific person the shot text names — its sex, age, build, skin and grooming must unmistakably match that person's profile above.

No text, captions, watermarks or annotations anywhere.

VISUAL APPROACH — this take (camera, framing, composition):
Frame from the entrance with the wind-rattled door edge acting as a narrow near-frame boundary, then draw the eye through the scattered lower-frame objects toward the two separated figures in the middle distance. Favor a gently compressed, observational lens feel rather than an expansive wide-angle distortion, with the room’s planes layered behind them. Keep the dusk light directional but natural, allowing soft falloff and restrained shadow to isolate their stillness within the ransacked emptiness.

PRIORITY OF INSTRUCTIONS: the VISUAL APPROACH above governs camera,
framing and composition ONLY. Every other instruction in this prompt —
location, time of day, people and their identity, poses, props,
carried state, physical rules, and the no-text rule — is binding
exactly as written earlier. If the VISUAL APPROACH appears to conflict
with any of them, the earlier instruction wins.

Edit this photograph. Apply ONLY the corrections listed below,
each in place, changing nothing else:

CORRECTIONS:
- Slide the background window pane open to match the previous shot.

PRESERVE EVERYTHING ELSE EXACTLY as in the original photograph —
every person, pose, face, garment, object, set feature, lighting,
camera and framing stay identical. No text, captions or
watermarks anywhere.

Create ONE FINAL photorealistic live-action film still of the moment below — contemporary South Korea, 2026; all people are Korean unless stated. TIME OF DAY (lock): dusk.

SHOT TEXT (authoritative, Korean): 눈을 크게 뜬 인우가 사색이 되어 굳어 있는 얼굴 클로즈업

LOCATION (lock): On the open deck of the small fishing boat at sunset, beside the lone passenger and her unfolded map and photograph. The shot takes place here — the attached PREVIOUS SHOT STILL shows this exact place.

CAMERA & FRAME (follow exactly — this is the composition authority for this still; the reference images supply identity and place, never the framing):
- CAMERA: At 인우’s eye level and slightly off his frontal axis, the static camera tightens to an extreme facial close-up that isolates his widened eyes and abruptly drained expression. His face occupies most of the frame with a slight three-quarter turn toward 수리영 off screen, while the deck falls into indistinct peripheral context.
- FRAMING SCALE: close-up
- FRAME LAYOUT: 인우 in the middle-center of the frame, foreground, looks toward 수리영 off screen.
- KEY BACKGROUND ELEMENTS: Fishing-boat deck (At sea) — Only a narrow, indistinct portion behind 인우 remains visible; used as Minimal blurred context at the margins of the facial close-up.
Compose the frame exactly as specified above — camera angle, subject scale and screen placement. Keep true physical scale between people and background elements: fixtures and distant objects occupy only the small screen area their real size and distance dictate; never enlarge a background object into a foreground presence.

THIS SHOT CONTINUES THE PREVIOUS SHOT: everything the attached still established — the place, its fixed features and wear, each person's clothing and state — persists. Any person in it who cannot move stays PRECISELY as photographed (body, pose, contact points, held objects); only the camera changes.

REALIZE FIGURATIVE LANGUAGE AS A LIVE-ACTION SHOT: the Korean shot
text may describe characters metaphorically, figuratively or with
exaggeration. Photograph what a real movie camera would actually
record on a physical set — exaggerated or figurative impressions
become realistic staging choices (lighting, distance, angle,
wardrobe), not literal fantasy imagery.
EVERY CHARACTER IS A HUMAN BEING: unless the story explicitly
features non-human or virtual beings (as in science fiction or
fantasy), every character — however indirectly, vaguely or
figuratively the text describes them — IS a real human. When the
text gives no direct visual description of a person, IMAGINE one
and still show them as a concrete, fully-formed human being:
always render the human form to the maximum extent — build,
posture, face, hands, clothing — never reduce a person to a shape,
blob, solid silhouette or abstract mass.

EXPRESSIONS ARE ACTED, NEVER ANATOMICAL: when the text describes a
person's eyes, face or presence emotionally or figuratively (vacant,
hollow, dazed, burning, lifeless gaze and the like), it describes an
ACTOR'S PERFORMANCE captured by a real camera — realize it ONLY
through gaze direction, focus, eyelids, facial muscles, stillness
and posture. Every living person's eyes remain anatomically normal
human eyes with a natural iris and pupil, natural sclera and normal
proportions; NEVER whiten, blank out, cloud over, glow, enlarge or
otherwise alter eyeballs, skin or anatomy — unless the story
explicitly declares that being non-human or supernatural in form.

PROPS FACE THE RIGHT WAY: every handheld or used object must be
oriented exactly as its real-world use requires. A person reading,
watching or operating something (a phone, a photograph, a paper,
any device) has its functional side — screen, front, page — facing
THEIR OWN eyes; the camera then sees whatever side the staging
geometry implies (often its back). Show the functional side to the
camera ONLY when the shot text itself stages it toward the viewer.
Never flip, mirror or reverse an object's front and back.

CARRIED STATE (persist exactly — must match the neighbouring shots of this scene): The old photograph remains in Su-ri-young’s hand or immediate possession, while her backpack and maps remain aboard the boat.

PEOPLE: the SHOT TEXT alone decides whether any person is visible in this shot. IF a person appears, they must be one of: 인우 (한국인 20대 중반 남성, 짧은 검은 머리, 반듯한 얼굴형, 젊은 인상) — never anyone else, and never add a person the shot text does not show. IF only part of a person is in frame (a hand, arm, foot, back, silhouette), that body part belongs to the specific person the shot text names — its sex, age, build, skin and grooming must unmistakably match that person's profile above.

No text, captions, watermarks or annotations anywhere.

Create ONE FINAL photorealistic live-action film still of the moment below — contemporary South Korea, 2026; all people are Korean unless stated. TIME OF DAY (lock): dusk.

SHOT TEXT (authoritative, Korean): 눈을 크게 뜬 인우가 사색이 되어 굳어 있는 얼굴 클로즈업

LOCATION (lock): On the open deck of the small fishing boat at sunset, beside the lone passenger and her unfolded map and photograph. The shot takes place here — the attached PREVIOUS SHOT STILL shows this exact place.

CAMERA & FRAME (follow exactly — this is the composition authority for this still; the reference images supply identity and place, never the framing):
- CAMERA: At 인우’s eye level and slightly off his frontal axis, the static camera tightens to an extreme facial close-up that isolates his widened eyes and abruptly drained expression. His face occupies most of the frame with a slight three-quarter turn toward 수리영 off screen, while the deck falls into indistinct peripheral context.
- FRAMING SCALE: close-up
- FRAME LAYOUT: 인우 in the middle-center of the frame, foreground, looks toward 수리영 off screen.
- KEY BACKGROUND ELEMENTS: Fishing-boat deck (At sea) — Only a narrow, indistinct portion behind 인우 remains visible; used as Minimal blurred context at the margins of the facial close-up.
Compose the frame exactly as specified above — camera angle, subject scale and screen placement. Keep true physical scale between people and background elements: fixtures and distant objects occupy only the small screen area their real size and distance dictate; never enlarge a background object into a foreground presence.

THIS SHOT CONTINUES THE PREVIOUS SHOT: everything the attached still established — the place, its fixed features and wear, each person's clothing and state — persists. Any person in it who cannot move stays PRECISELY as photographed (body, pose, contact points, held objects); only the camera changes.

REALIZE FIGURATIVE LANGUAGE AS A LIVE-ACTION SHOT: the Korean shot
text may describe characters metaphorically, figuratively or with
exaggeration. Photograph what a real movie camera would actually
record on a physical set — exaggerated or figurative impressions
become realistic staging choices (lighting, distance, angle,
wardrobe), not literal fantasy imagery.
EVERY CHARACTER IS A HUMAN BEING: unless the story explicitly
features non-human or virtual beings (as in science fiction or
fantasy), every character — however indirectly, vaguely or
figuratively the text describes them — IS a real human. When the
text gives no direct visual description of a person, IMAGINE one
and still show them as a concrete, fully-formed human being:
always render the human form to the maximum extent — build,
posture, face, hands, clothing — never reduce a person to a shape,
blob, solid silhouette or abstract mass.

EXPRESSIONS ARE ACTED, NEVER ANATOMICAL: when the text describes a
person's eyes, face or presence emotionally or figuratively (vacant,
hollow, dazed, burning, lifeless gaze and the like), it describes an
ACTOR'S PERFORMANCE captured by a real camera — realize it ONLY
through gaze direction, focus, eyelids, facial muscles, stillness
and posture. Every living person's eyes remain anatomically normal
human eyes with a natural iris and pupil, natural sclera and normal
proportions; NEVER whiten, blank out, cloud over, glow, enlarge or
otherwise alter eyeballs, skin or anatomy — unless the story
explicitly declares that being non-human or supernatural in form.

PROPS FACE THE RIGHT WAY: every handheld or used object must be
oriented exactly as its real-world use requires. A person reading,
watching or operating something (a phone, a photograph, a paper,
any device) has its functional side — screen, front, page — facing
THEIR OWN eyes; the camera then sees whatever side the staging
geometry implies (often its back). Show the functional side to the
camera ONLY when the shot text itself stages it toward the viewer.
Never flip, mirror or reverse an object's front and back.

CARRIED STATE (persist exactly — must match the neighbouring shots of this scene): The old photograph remains in Su-ri-young’s hand or immediate possession, while her backpack and maps remain aboard the boat.

PEOPLE: the SHOT TEXT alone decides whether any person is visible in this shot. IF a person appears, they must be one of: 인우 (한국인 20대 중반 남성, 짧은 검은 머리, 반듯한 얼굴형, 젊은 인상) — never anyone else, and never add a person the shot text does not show. IF only part of a person is in frame (a hand, arm, foot, back, silhouette), that body part belongs to the specific person the shot text names — its sex, age, build, skin and grooming must unmistakably match that person's profile above.

No text, captions, watermarks or annotations anywhere.

VISUAL APPROACH — this take (camera, framing, composition):
Use a compressed, claustrophobic lens feel with extremely shallow depth of field, holding both widened eyes in crisp focus while the rest of his face falls off gently and the deck dissolves at the margins. Let the last low sunset light graze one side of his face, opposed by soft cool dusk ambience, emphasizing locked facial muscles and a suddenly bloodless impression through natural performance and color contrast.

PRIORITY OF INSTRUCTIONS: the VISUAL APPROACH above governs camera,
framing and composition ONLY. Every other instruction in this prompt —
location, time of day, people and their identity, poses, props,
carried state, physical rules, and the no-text rule — is binding
exactly as written earlier. If the VISUAL APPROACH appears to conflict
with any of them, the earlier instruction wins.

Create ONE FINAL photorealistic live-action film still of the moment below — contemporary South Korea, 2026; all people are Korean unless stated. TIME OF DAY (lock): dusk.

SHOT TEXT (authoritative, Korean): 눈을 크게 뜬 인우가 사색이 되어 굳어 있는 얼굴 클로즈업

LOCATION (lock): On the open deck of the small fishing boat at sunset, beside the lone passenger and her unfolded map and photograph. The shot takes place here — the attached PREVIOUS SHOT STILL shows this exact place.

CAMERA & FRAME (follow exactly — this is the composition authority for this still; the reference images supply identity and place, never the framing):
- CAMERA: At 인우’s eye level and slightly off his frontal axis, the static camera tightens to an extreme facial close-up that isolates his widened eyes and abruptly drained expression. His face occupies most of the frame with a slight three-quarter turn toward 수리영 off screen, while the deck falls into indistinct peripheral context.
- FRAMING SCALE: close-up
- FRAME LAYOUT: 인우 in the middle-center of the frame, foreground, looks toward 수리영 off screen.
- KEY BACKGROUND ELEMENTS: Fishing-boat deck (At sea) — Only a narrow, indistinct portion behind 인우 remains visible; used as Minimal blurred context at the margins of the facial close-up.
Compose the frame exactly as specified above — camera angle, subject scale and screen placement. Keep true physical scale between people and background elements: fixtures and distant objects occupy only the small screen area their real size and distance dictate; never enlarge a background object into a foreground presence.

THIS SHOT CONTINUES THE PREVIOUS SHOT: everything the attached still established — the place, its fixed features and wear, each person's clothing and state — persists. Any person in it who cannot move stays PRECISELY as photographed (body, pose, contact points, held objects); only the camera changes.

REALIZE FIGURATIVE LANGUAGE AS A LIVE-ACTION SHOT: the Korean shot
text may describe characters metaphorically, figuratively or with
exaggeration. Photograph what a real movie camera would actually
record on a physical set — exaggerated or figurative impressions
become realistic staging choices (lighting, distance, angle,
wardrobe), not literal fantasy imagery.
EVERY CHARACTER IS A HUMAN BEING: unless the story explicitly
features non-human or virtual beings (as in science fiction or
fantasy), every character — however indirectly, vaguely or
figuratively the text describes them — IS a real human. When the
text gives no direct visual description of a person, IMAGINE one
and still show them as a concrete, fully-formed human being:
always render the human form to the maximum extent — build,
posture, face, hands, clothing — never reduce a person to a shape,
blob, solid silhouette or abstract mass.

EXPRESSIONS ARE ACTED, NEVER ANATOMICAL: when the text describes a
person's eyes, face or presence emotionally or figuratively (vacant,
hollow, dazed, burning, lifeless gaze and the like), it describes an
ACTOR'S PERFORMANCE captured by a real camera — realize it ONLY
through gaze direction, focus, eyelids, facial muscles, stillness
and posture. Every living person's eyes remain anatomically normal
human eyes with a natural iris and pupil, natural sclera and normal
proportions; NEVER whiten, blank out, cloud over, glow, enlarge or
otherwise alter eyeballs, skin or anatomy — unless the story
explicitly declares that being non-human or supernatural in form.

PROPS FACE THE RIGHT WAY: every handheld or used object must be
oriented exactly as its real-world use requires. A person reading,
watching or operating something (a phone, a photograph, a paper,
any device) has its functional side — screen, front, page — facing
THEIR OWN eyes; the camera then sees whatever side the staging
geometry implies (often its back). Show the functional side to the
camera ONLY when the shot text itself stages it toward the viewer.
Never flip, mirror or reverse an object's front and back.

CARRIED STATE (persist exactly — must match the neighbouring shots of this scene): The old photograph remains in Su-ri-young’s hand or immediate possession, while her backpack and maps remain aboard the boat.

PEOPLE: the SHOT TEXT alone decides whether any person is visible in this shot. IF a person appears, they must be one of: 인우 (한국인 20대 중반 남성, 짧은 검은 머리, 반듯한 얼굴형, 젊은 인상) — never anyone else, and never add a person the shot text does not show. IF only part of a person is in frame (a hand, arm, foot, back, silhouette), that body part belongs to the specific person the shot text names — its sex, age, build, skin and grooming must unmistakably match that person's profile above.

No text, captions, watermarks or annotations anywhere.

VISUAL APPROACH — this take (camera, framing, composition):
Photograph from an intimately close, natural-perspective viewpoint that gives the face immediate physical presence without stylization, with enough depth to retain fine tension around the eyes, brow, and jaw. Favor soft, even dusk illumination with only a faint warm directional trace from sunset; keep contrast restrained and the peripheral deck subtly legible, making his rigid stillness and fixed off-screen gaze feel starkly real.

PRIORITY OF INSTRUCTIONS: the VISUAL APPROACH above governs camera,
framing and composition ONLY. Every other instruction in this prompt —
location, time of day, people and their identity, poses, props,
carried state, physical rules, and the no-text rule — is binding
exactly as written earlier. If the VISUAL APPROACH appears to conflict
with any of them, the earlier instruction wins.

Edit this photograph. Apply ONLY the corrections listed below,
each in place, changing nothing else:

CORRECTIONS:
- Add the grey knit beanie to In-woo's head exactly as worn in his character reference image.

PRESERVE EVERYTHING ELSE EXACTLY as in the original photograph —
every person, pose, face, garment, object, set feature, lighting,
camera and framing stay identical. No text, captions or
watermarks anywhere.

Create ONE FINAL photorealistic live-action film still of the moment below — contemporary South Korea, 2026; all people are Korean unless stated. TIME OF DAY (lock): dusk.

SHOT TEXT (authoritative, Korean): 창백해진 인우가 입을 반쯤 벌린 채 충격에 빠진 얼굴 정면 클로즈업

LOCATION (lock): Inside the small fishing boat's wheelhouse, at the helm with sunset light entering from the windows. The shot takes place here — the attached PREVIOUS SHOT STILL shows this exact place.

CAMERA & FRAME (follow exactly — this is the composition authority for this still; the reference images supply identity and place, never the framing):
- CAMERA: At eye level and directly in front of 인우, the static camera holds a close portrait as his pale face occupies most of the center frame, his parted mouth slightly below the horizontal midpoint. The frontal placement functions as a moment of involuntary confrontation, while his unfocused eyes look beyond the lens rather than addressing it.
- FRAMING SCALE: close-up
- FRAME LAYOUT: 인우 in the middle-center of the frame, midground.
- KEY BACKGROUND ELEMENTS: wheelhouse interior (Mostly excluded by the close framing); used as Soft, restricted background context that keeps attention on 인우's shock before the following reveal.
Compose the frame exactly as specified above — camera angle, subject scale and screen placement. Keep true physical scale between people and background elements: fixtures and distant objects occupy only the small screen area their real size and distance dictate; never enlarge a background object into a foreground presence.

THIS SHOT CONTINUES THE PREVIOUS SHOT: everything the attached still established — the place, its fixed features and wear, each person's clothing and state — persists. Any person in it who cannot move stays PRECISELY as photographed (body, pose, contact points, held objects); only the camera changes.

REALIZE FIGURATIVE LANGUAGE AS A LIVE-ACTION SHOT: the Korean shot
text may describe characters metaphorically, figuratively or with
exaggeration. Photograph what a real movie camera would actually
record on a physical set — exaggerated or figurative impressions
become realistic staging choices (lighting, distance, angle,
wardrobe), not literal fantasy imagery.
EVERY CHARACTER IS A HUMAN BEING: unless the story explicitly
features non-human or virtual beings (as in science fiction or
fantasy), every character — however indirectly, vaguely or
figuratively the text describes them — IS a real human. When the
text gives no direct visual description of a person, IMAGINE one
and still show them as a concrete, fully-formed human being:
always render the human form to the maximum extent — build,
posture, face, hands, clothing — never reduce a person to a shape,
blob, solid silhouette or abstract mass.

EXPRESSIONS ARE ACTED, NEVER ANATOMICAL: when the text describes a
person's eyes, face or presence emotionally or figuratively (vacant,
hollow, dazed, burning, lifeless gaze and the like), it describes an
ACTOR'S PERFORMANCE captured by a real camera — realize it ONLY
through gaze direction, focus, eyelids, facial muscles, stillness
and posture. Every living person's eyes remain anatomically normal
human eyes with a natural iris and pupil, natural sclera and normal
proportions; NEVER whiten, blank out, cloud over, glow, enlarge or
otherwise alter eyeballs, skin or anatomy — unless the story
explicitly declares that being non-human or supernatural in form.

PROPS FACE THE RIGHT WAY: every handheld or used object must be
oriented exactly as its real-world use requires. A person reading,
watching or operating something (a phone, a photograph, a paper,
any device) has its functional side — screen, front, page — facing
THEIR OWN eyes; the camera then sees whatever side the staging
geometry implies (often its back). Show the functional side to the
camera ONLY when the shot text itself stages it toward the viewer.
Never flip, mirror or reverse an object's front and back.

CARRIED STATE (persist exactly — must match the neighbouring shots of this scene): Su-ri-young still has the old photograph identifying the child as Min-suk; her backpack, maps, coordinate note, and phone remain aboard the boat.

PEOPLE: the SHOT TEXT alone decides whether any person is visible in this shot. IF a person appears, they must be one of: 인우 (한국인 20대 중반 남성, 짧은 검은 머리, 반듯한 얼굴형, 젊은 인상) — never anyone else, and never add a person the shot text does not show. IF only part of a person is in frame (a hand, arm, foot, back, silhouette), that body part belongs to the specific person the shot text names — its sex, age, build, skin and grooming must unmistakably match that person's profile above.

No text, captions, watermarks or annotations anywhere.

Create ONE FINAL photorealistic live-action film still of the moment below — contemporary South Korea, 2026; all people are Korean unless stated. TIME OF DAY (lock): dusk.

SHOT TEXT (authoritative, Korean): 창백해진 인우가 입을 반쯤 벌린 채 충격에 빠진 얼굴 정면 클로즈업

LOCATION (lock): Inside the small fishing boat's wheelhouse, at the helm with sunset light entering from the windows. The shot takes place here — the attached PREVIOUS SHOT STILL shows this exact place.

CAMERA & FRAME (follow exactly — this is the composition authority for this still; the reference images supply identity and place, never the framing):
- CAMERA: At eye level and directly in front of 인우, the static camera holds a close portrait as his pale face occupies most of the center frame, his parted mouth slightly below the horizontal midpoint. The frontal placement functions as a moment of involuntary confrontation, while his unfocused eyes look beyond the lens rather than addressing it.
- FRAMING SCALE: close-up
- FRAME LAYOUT: 인우 in the middle-center of the frame, midground.
- KEY BACKGROUND ELEMENTS: wheelhouse interior (Mostly excluded by the close framing); used as Soft, restricted background context that keeps attention on 인우's shock before the following reveal.
Compose the frame exactly as specified above — camera angle, subject scale and screen placement. Keep true physical scale between people and background elements: fixtures and distant objects occupy only the small screen area their real size and distance dictate; never enlarge a background object into a foreground presence.

THIS SHOT CONTINUES THE PREVIOUS SHOT: everything the attached still established — the place, its fixed features and wear, each person's clothing and state — persists. Any person in it who cannot move stays PRECISELY as photographed (body, pose, contact points, held objects); only the camera changes.

REALIZE FIGURATIVE LANGUAGE AS A LIVE-ACTION SHOT: the Korean shot
text may describe characters metaphorically, figuratively or with
exaggeration. Photograph what a real movie camera would actually
record on a physical set — exaggerated or figurative impressions
become realistic staging choices (lighting, distance, angle,
wardrobe), not literal fantasy imagery.
EVERY CHARACTER IS A HUMAN BEING: unless the story explicitly
features non-human or virtual beings (as in science fiction or
fantasy), every character — however indirectly, vaguely or
figuratively the text describes them — IS a real human. When the
text gives no direct visual description of a person, IMAGINE one
and still show them as a concrete, fully-formed human being:
always render the human form to the maximum extent — build,
posture, face, hands, clothing — never reduce a person to a shape,
blob, solid silhouette or abstract mass.

EXPRESSIONS ARE ACTED, NEVER ANATOMICAL: when the text describes a
person's eyes, face or presence emotionally or figuratively (vacant,
hollow, dazed, burning, lifeless gaze and the like), it describes an
ACTOR'S PERFORMANCE captured by a real camera — realize it ONLY
through gaze direction, focus, eyelids, facial muscles, stillness
and posture. Every living person's eyes remain anatomically normal
human eyes with a natural iris and pupil, natural sclera and normal
proportions; NEVER whiten, blank out, cloud over, glow, enlarge or
otherwise alter eyeballs, skin or anatomy — unless the story
explicitly declares that being non-human or supernatural in form.

PROPS FACE THE RIGHT WAY: every handheld or used object must be
oriented exactly as its real-world use requires. A person reading,
watching or operating something (a phone, a photograph, a paper,
any device) has its functional side — screen, front, page — facing
THEIR OWN eyes; the camera then sees whatever side the staging
geometry implies (often its back). Show the functional side to the
camera ONLY when the shot text itself stages it toward the viewer.
Never flip, mirror or reverse an object's front and back.

CARRIED STATE (persist exactly — must match the neighbouring shots of this scene): Su-ri-young still has the old photograph identifying the child as Min-suk; her backpack, maps, coordinate note, and phone remain aboard the boat.

PEOPLE: the SHOT TEXT alone decides whether any person is visible in this shot. IF a person appears, they must be one of: 인우 (한국인 20대 중반 남성, 짧은 검은 머리, 반듯한 얼굴형, 젊은 인상) — never anyone else, and never add a person the shot text does not show. IF only part of a person is in frame (a hand, arm, foot, back, silhouette), that body part belongs to the specific person the shot text names — its sex, age, build, skin and grooming must unmistakably match that person's profile above.

No text, captions, watermarks or annotations anywhere.

VISUAL APPROACH — this take (camera, framing, composition):
Use a compressed long-lens portrait feel with extremely shallow depth of field, holding the face in precise focus while the restricted wheelhouse context dissolves into soft dusk color. Let the low sunset illumination skim gently from one side, leaving restrained interior shadow across the opposite side; keep the exposure subdued and natural, emphasizing stillness, pallor, and the minute tension around the unfocused eyes and parted lips.

PRIORITY OF INSTRUCTIONS: the VISUAL APPROACH above governs camera,
framing and composition ONLY. Every other instruction in this prompt —
location, time of day, people and their identity, poses, props,
carried state, physical rules, and the no-text rule — is binding
exactly as written earlier. If the VISUAL APPROACH appears to conflict
with any of them, the earlier instruction wins.

Create ONE FINAL photorealistic live-action film still of the moment below — contemporary South Korea, 2026; all people are Korean unless stated. TIME OF DAY (lock): dusk.

SHOT TEXT (authoritative, Korean): 창백해진 인우가 입을 반쯤 벌린 채 충격에 빠진 얼굴 정면 클로즈업

LOCATION (lock): Inside the small fishing boat's wheelhouse, at the helm with sunset light entering from the windows. The shot takes place here — the attached PREVIOUS SHOT STILL shows this exact place.

CAMERA & FRAME (follow exactly — this is the composition authority for this still; the reference images supply identity and place, never the framing):
- CAMERA: At eye level and directly in front of 인우, the static camera holds a close portrait as his pale face occupies most of the center frame, his parted mouth slightly below the horizontal midpoint. The frontal placement functions as a moment of involuntary confrontation, while his unfocused eyes look beyond the lens rather than addressing it.
- FRAMING SCALE: close-up
- FRAME LAYOUT: 인우 in the middle-center of the frame, midground.
- KEY BACKGROUND ELEMENTS: wheelhouse interior (Mostly excluded by the close framing); used as Soft, restricted background context that keeps attention on 인우's shock before the following reveal.
Compose the frame exactly as specified above — camera angle, subject scale and screen placement. Keep true physical scale between people and background elements: fixtures and distant objects occupy only the small screen area their real size and distance dictate; never enlarge a background object into a foreground presence.

THIS SHOT CONTINUES THE PREVIOUS SHOT: everything the attached still established — the place, its fixed features and wear, each person's clothing and state — persists. Any person in it who cannot move stays PRECISELY as photographed (body, pose, contact points, held objects); only the camera changes.

REALIZE FIGURATIVE LANGUAGE AS A LIVE-ACTION SHOT: the Korean shot
text may describe characters metaphorically, figuratively or with
exaggeration. Photograph what a real movie camera would actually
record on a physical set — exaggerated or figurative impressions
become realistic staging choices (lighting, distance, angle,
wardrobe), not literal fantasy imagery.
EVERY CHARACTER IS A HUMAN BEING: unless the story explicitly
features non-human or virtual beings (as in science fiction or
fantasy), every character — however indirectly, vaguely or
figuratively the text describes them — IS a real human. When the
text gives no direct visual description of a person, IMAGINE one
and still show them as a concrete, fully-formed human being:
always render the human form to the maximum extent — build,
posture, face, hands, clothing — never reduce a person to a shape,
blob, solid silhouette or abstract mass.

EXPRESSIONS ARE ACTED, NEVER ANATOMICAL: when the text describes a
person's eyes, face or presence emotionally or figuratively (vacant,
hollow, dazed, burning, lifeless gaze and the like), it describes an
ACTOR'S PERFORMANCE captured by a real camera — realize it ONLY
through gaze direction, focus, eyelids, facial muscles, stillness
and posture. Every living person's eyes remain anatomically normal
human eyes with a natural iris and pupil, natural sclera and normal
proportions; NEVER whiten, blank out, cloud over, glow, enlarge or
otherwise alter eyeballs, skin or anatomy — unless the story
explicitly declares that being non-human or supernatural in form.

PROPS FACE THE RIGHT WAY: every handheld or used object must be
oriented exactly as its real-world use requires. A person reading,
watching or operating something (a phone, a photograph, a paper,
any device) has its functional side — screen, front, page — facing
THEIR OWN eyes; the camera then sees whatever side the staging
geometry implies (often its back). Show the functional side to the
camera ONLY when the shot text itself stages it toward the viewer.
Never flip, mirror or reverse an object's front and back.

CARRIED STATE (persist exactly — must match the neighbouring shots of this scene): Su-ri-young still has the old photograph identifying the child as Min-suk; her backpack, maps, coordinate note, and phone remain aboard the boat.

PEOPLE: the SHOT TEXT alone decides whether any person is visible in this shot. IF a person appears, they must be one of: 인우 (한국인 20대 중반 남성, 짧은 검은 머리, 반듯한 얼굴형, 젊은 인상) — never anyone else, and never add a person the shot text does not show. IF only part of a person is in frame (a hand, arm, foot, back, silhouette), that body part belongs to the specific person the shot text names — its sex, age, build, skin and grooming must unmistakably match that person's profile above.

No text, captions, watermarks or annotations anywhere.

VISUAL APPROACH — this take (camera, framing, composition):
Use a natural-perspective close portrait with moderate depth, preserving a faint sense of the wheelhouse geometry behind him without competing with the face. Frame with severe frontal balance and minimal visual softness, allowing the dusk window light to fall broadly and evenly before fading into the dim interior; favor quiet tonal realism, fine facial texture, and an unembellished documentary immediacy.

PRIORITY OF INSTRUCTIONS: the VISUAL APPROACH above governs camera,
framing and composition ONLY. Every other instruction in this prompt —
location, time of day, people and their identity, poses, props,
carried state, physical rules, and the no-text rule — is binding
exactly as written earlier. If the VISUAL APPROACH appears to conflict
with any of them, the earlier instruction wins.

Edit this photograph. Apply ONLY the corrections listed below,
each in place, changing nothing else:

CORRECTIONS:
- Adjust In-woo's skin tone to be noticeably pale and drained of color, removing the warm flush.
- Shift In-woo's gaze slightly off-center and soften his focus so he looks blankly beyond the lens rather than making direct eye contact.

PRESERVE EVERYTHING ELSE EXACTLY as in the original photograph —
every person, pose, face, garment, object, set feature, lighting,
camera and framing stay identical. No text, captions or
watermarks anywhere.

Create ONE FINAL photorealistic live-action film still of the moment below — contemporary South Korea, 2026; all people are Korean unless stated. TIME OF DAY (lock): dusk.

SHOT TEXT (authoritative, Korean): 인우의 어깨너머로 조타실 벽면에 빼곡하게 그려진 붉은색 원 형태들이 보이는 구도

LOCATION (lock): Inside the boat's wheelhouse, facing a wall densely covered with red circular marks in the fading sunset light. The shot takes place here — the attached PREVIOUS SHOT STILL shows this exact place.

CAMERA & FRAME (follow exactly — this is the composition authority for this still; the reference images supply identity and place, never the framing):
- CAMERA: The lateral track finishes at eye level just behind 인우's shoulder, preserving his rigid foreground silhouette while redirecting the composition toward the marked wall. His shoulder occupies the left edge, and the numerous circular forms spread across the center and right background, with the red circle held as the sharpest crimson accent before the camera pans onward.
- FRAMING SCALE: medium shot
- FRAME LAYOUT: wall covered with circular forms in the middle-center of the frame, background; 인우's shoulder in the middle-left of the frame, foreground.
- KEY BACKGROUND ELEMENTS: wheelhouse wall with circular forms (Densely covered with circular forms) — The marked face of the wall is turned toward the camera, showing numerous circular forms and a newly discovered red circle; used as Primary reveal beyond 인우's shoulder, distributing repeated forms across the background plane.
Compose the frame exactly as specified above — camera angle, subject scale and screen placement. Keep true physical scale between people and background elements: fixtures and distant objects occupy only the small screen area their real size and distance dictate; never enlarge a background object into a foreground presence.

THIS SHOT CONTINUES THE PREVIOUS SHOT: everything the attached still established — the place, its fixed features and wear, each person's clothing and state — persists. Any person in it who cannot move stays PRECISELY as photographed (body, pose, contact points, held objects); only the camera changes.

REALIZE FIGURATIVE LANGUAGE AS A LIVE-ACTION SHOT: the Korean shot
text may describe characters metaphorically, figuratively or with
exaggeration. Photograph what a real movie camera would actually
record on a physical set — exaggerated or figurative impressions
become realistic staging choices (lighting, distance, angle,
wardrobe), not literal fantasy imagery.
EVERY CHARACTER IS A HUMAN BEING: unless the story explicitly
features non-human or virtual beings (as in science fiction or
fantasy), every character — however indirectly, vaguely or
figuratively the text describes them — IS a real human. When the
text gives no direct visual description of a person, IMAGINE one
and still show them as a concrete, fully-formed human being:
always render the human form to the maximum extent — build,
posture, face, hands, clothing — never reduce a person to a shape,
blob, solid silhouette or abstract mass.

EXPRESSIONS ARE ACTED, NEVER ANATOMICAL: when the text describes a
person's eyes, face or presence emotionally or figuratively (vacant,
hollow, dazed, burning, lifeless gaze and the like), it describes an
ACTOR'S PERFORMANCE captured by a real camera — realize it ONLY
through gaze direction, focus, eyelids, facial muscles, stillness
and posture. Every living person's eyes remain anatomically normal
human eyes with a natural iris and pupil, natural sclera and normal
proportions; NEVER whiten, blank out, cloud over, glow, enlarge or
otherwise alter eyeballs, skin or anatomy — unless the story
explicitly declares that being non-human or supernatural in form.

PROPS FACE THE RIGHT WAY: every handheld or used object must be
oriented exactly as its real-world use requires. A person reading,
watching or operating something (a phone, a photograph, a paper,
any device) has its functional side — screen, front, page — facing
THEIR OWN eyes; the camera then sees whatever side the staging
geometry implies (often its back). Show the functional side to the
camera ONLY when the shot text itself stages it toward the viewer.
Never flip, mirror or reverse an object's front and back.

CARRIED STATE (persist exactly — must match the neighbouring shots of this scene): Numerous circular marks, including a red circle, remain densely visible on the wheelhouse walls. Su-ri-young still has the old photograph with her, and her backpack and maps remain aboard.

PEOPLE: the SHOT TEXT alone decides whether any person is visible in this shot. IF a person appears, they must be one of: 인우 (한국인 20대 중반 남성, 짧은 검은 머리, 반듯한 얼굴형, 젊은 인상) — never anyone else, and never add a person the shot text does not show. IF only part of a person is in frame (a hand, arm, foot, back, silhouette), that body part belongs to the specific person the shot text names — its sex, age, build, skin and grooming must unmistakably match that person's profile above.

No text, captions, watermarks or annotations anywhere.

Create ONE FINAL photorealistic live-action film still of the moment below — contemporary South Korea, 2026; all people are Korean unless stated. TIME OF DAY (lock): dusk.

SHOT TEXT (authoritative, Korean): 인우의 어깨너머로 조타실 벽면에 빼곡하게 그려진 붉은색 원 형태들이 보이는 구도

LOCATION (lock): Inside the boat's wheelhouse, facing a wall densely covered with red circular marks in the fading sunset light. The shot takes place here — the attached PREVIOUS SHOT STILL shows this exact place.

CAMERA & FRAME (follow exactly — this is the composition authority for this still; the reference images supply identity and place, never the framing):
- CAMERA: The lateral track finishes at eye level just behind 인우's shoulder, preserving his rigid foreground silhouette while redirecting the composition toward the marked wall. His shoulder occupies the left edge, and the numerous circular forms spread across the center and right background, with the red circle held as the sharpest crimson accent before the camera pans onward.
- FRAMING SCALE: medium shot
- FRAME LAYOUT: wall covered with circular forms in the middle-center of the frame, background; 인우's shoulder in the middle-left of the frame, foreground.
- KEY BACKGROUND ELEMENTS: wheelhouse wall with circular forms (Densely covered with circular forms) — The marked face of the wall is turned toward the camera, showing numerous circular forms and a newly discovered red circle; used as Primary reveal beyond 인우's shoulder, distributing repeated forms across the background plane.
Compose the frame exactly as specified above — camera angle, subject scale and screen placement. Keep true physical scale between people and background elements: fixtures and distant objects occupy only the small screen area their real size and distance dictate; never enlarge a background object into a foreground presence.

THIS SHOT CONTINUES THE PREVIOUS SHOT: everything the attached still established — the place, its fixed features and wear, each person's clothing and state — persists. Any person in it who cannot move stays PRECISELY as photographed (body, pose, contact points, held objects); only the camera changes.

REALIZE FIGURATIVE LANGUAGE AS A LIVE-ACTION SHOT: the Korean shot
text may describe characters metaphorically, figuratively or with
exaggeration. Photograph what a real movie camera would actually
record on a physical set — exaggerated or figurative impressions
become realistic staging choices (lighting, distance, angle,
wardrobe), not literal fantasy imagery.
EVERY CHARACTER IS A HUMAN BEING: unless the story explicitly
features non-human or virtual beings (as in science fiction or
fantasy), every character — however indirectly, vaguely or
figuratively the text describes them — IS a real human. When the
text gives no direct visual description of a person, IMAGINE one
and still show them as a concrete, fully-formed human being:
always render the human form to the maximum extent — build,
posture, face, hands, clothing — never reduce a person to a shape,
blob, solid silhouette or abstract mass.

EXPRESSIONS ARE ACTED, NEVER ANATOMICAL: when the text describes a
person's eyes, face or presence emotionally or figuratively (vacant,
hollow, dazed, burning, lifeless gaze and the like), it describes an
ACTOR'S PERFORMANCE captured by a real camera — realize it ONLY
through gaze direction, focus, eyelids, facial muscles, stillness
and posture. Every living person's eyes remain anatomically normal
human eyes with a natural iris and pupil, natural sclera and normal
proportions; NEVER whiten, blank out, cloud over, glow, enlarge or
otherwise alter eyeballs, skin or anatomy — unless the story
explicitly declares that being non-human or supernatural in form.

PROPS FACE THE RIGHT WAY: every handheld or used object must be
oriented exactly as its real-world use requires. A person reading,
watching or operating something (a phone, a photograph, a paper,
any device) has its functional side — screen, front, page — facing
THEIR OWN eyes; the camera then sees whatever side the staging
geometry implies (often its back). Show the functional side to the
camera ONLY when the shot text itself stages it toward the viewer.
Never flip, mirror or reverse an object's front and back.

CARRIED STATE (persist exactly — must match the neighbouring shots of this scene): Numerous circular marks, including a red circle, remain densely visible on the wheelhouse walls. Su-ri-young still has the old photograph with her, and her backpack and maps remain aboard.

PEOPLE: the SHOT TEXT alone decides whether any person is visible in this shot. IF a person appears, they must be one of: 인우 (한국인 20대 중반 남성, 짧은 검은 머리, 반듯한 얼굴형, 젊은 인상) — never anyone else, and never add a person the shot text does not show. IF only part of a person is in frame (a hand, arm, foot, back, silhouette), that body part belongs to the specific person the shot text names — its sex, age, build, skin and grooming must unmistakably match that person's profile above.

No text, captions, watermarks or annotations anywhere.

VISUAL APPROACH — this take (camera, framing, composition):
Use a restrained, compressed lens feel with deep focus, holding the eye-level medium framing in severe stillness. Keep the shoulder as a dark, rigid left-edge foreground anchor while the marked wall reads as a dense, nearly frontal field across the center and right; let the fading dusk light fall softly and laterally, preserving the newly discovered red circle as the crispest crimson accent amid the repeated forms.

PRIORITY OF INSTRUCTIONS: the VISUAL APPROACH above governs camera,
framing and composition ONLY. Every other instruction in this prompt —
location, time of day, people and their identity, poses, props,
carried state, physical rules, and the no-text rule — is binding
exactly as written earlier. If the VISUAL APPROACH appears to conflict
with any of them, the earlier instruction wins.

Create ONE FINAL photorealistic live-action film still of the moment below — contemporary South Korea, 2026; all people are Korean unless stated. TIME OF DAY (lock): dusk.

SHOT TEXT (authoritative, Korean): 인우의 어깨너머로 조타실 벽면에 빼곡하게 그려진 붉은색 원 형태들이 보이는 구도

LOCATION (lock): Inside the boat's wheelhouse, facing a wall densely covered with red circular marks in the fading sunset light. The shot takes place here — the attached PREVIOUS SHOT STILL shows this exact place.

CAMERA & FRAME (follow exactly — this is the composition authority for this still; the reference images supply identity and place, never the framing):
- CAMERA: The lateral track finishes at eye level just behind 인우's shoulder, preserving his rigid foreground silhouette while redirecting the composition toward the marked wall. His shoulder occupies the left edge, and the numerous circular forms spread across the center and right background, with the red circle held as the sharpest crimson accent before the camera pans onward.
- FRAMING SCALE: medium shot
- FRAME LAYOUT: wall covered with circular forms in the middle-center of the frame, background; 인우's shoulder in the middle-left of the frame, foreground.
- KEY BACKGROUND ELEMENTS: wheelhouse wall with circular forms (Densely covered with circular forms) — The marked face of the wall is turned toward the camera, showing numerous circular forms and a newly discovered red circle; used as Primary reveal beyond 인우's shoulder, distributing repeated forms across the background plane.
Compose the frame exactly as specified above — camera angle, subject scale and screen placement. Keep true physical scale between people and background elements: fixtures and distant objects occupy only the small screen area their real size and distance dictate; never enlarge a background object into a foreground presence.

THIS SHOT CONTINUES THE PREVIOUS SHOT: everything the attached still established — the place, its fixed features and wear, each person's clothing and state — persists. Any person in it who cannot move stays PRECISELY as photographed (body, pose, contact points, held objects); only the camera changes.

REALIZE FIGURATIVE LANGUAGE AS A LIVE-ACTION SHOT: the Korean shot
text may describe characters metaphorically, figuratively or with
exaggeration. Photograph what a real movie camera would actually
record on a physical set — exaggerated or figurative impressions
become realistic staging choices (lighting, distance, angle,
wardrobe), not literal fantasy imagery.
EVERY CHARACTER IS A HUMAN BEING: unless the story explicitly
features non-human or virtual beings (as in science fiction or
fantasy), every character — however indirectly, vaguely or
figuratively the text describes them — IS a real human. When the
text gives no direct visual description of a person, IMAGINE one
and still show them as a concrete, fully-formed human being:
always render the human form to the maximum extent — build,
posture, face, hands, clothing — never reduce a person to a shape,
blob, solid silhouette or abstract mass.

EXPRESSIONS ARE ACTED, NEVER ANATOMICAL: when the text describes a
person's eyes, face or presence emotionally or figuratively (vacant,
hollow, dazed, burning, lifeless gaze and the like), it describes an
ACTOR'S PERFORMANCE captured by a real camera — realize it ONLY
through gaze direction, focus, eyelids, facial muscles, stillness
and posture. Every living person's eyes remain anatomically normal
human eyes with a natural iris and pupil, natural sclera and normal
proportions; NEVER whiten, blank out, cloud over, glow, enlarge or
otherwise alter eyeballs, skin or anatomy — unless the story
explicitly declares that being non-human or supernatural in form.

PROPS FACE THE RIGHT WAY: every handheld or used object must be
oriented exactly as its real-world use requires. A person reading,
watching or operating something (a phone, a photograph, a paper,
any device) has its functional side — screen, front, page — facing
THEIR OWN eyes; the camera then sees whatever side the staging
geometry implies (often its back). Show the functional side to the
camera ONLY when the shot text itself stages it toward the viewer.
Never flip, mirror or reverse an object's front and back.

CARRIED STATE (persist exactly — must match the neighbouring shots of this scene): Numerous circular marks, including a red circle, remain densely visible on the wheelhouse walls. Su-ri-young still has the old photograph with her, and her backpack and maps remain aboard.

PEOPLE: the SHOT TEXT alone decides whether any person is visible in this shot. IF a person appears, they must be one of: 인우 (한국인 20대 중반 남성, 짧은 검은 머리, 반듯한 얼굴형, 젊은 인상) — never anyone else, and never add a person the shot text does not show. IF only part of a person is in frame (a hand, arm, foot, back, silhouette), that body part belongs to the specific person the shot text names — its sex, age, build, skin and grooming must unmistakably match that person's profile above.

No text, captions, watermarks or annotations anywhere.

VISUAL APPROACH — this take (camera, framing, composition):
Photograph the reveal with pronounced foreground-to-background separation: render the near shoulder softly out of focus at the middle-left while snapping the wall and its red circular forms into precise focus across the center and right. Use a slightly broader, more spatial lens feel at eye level so the over-shoulder depth is palpable, with low dusk illumination grazing the wall’s surface and making the sharp red circle command attention without brightening the surrounding wheelhouse unnaturally.

PRIORITY OF INSTRUCTIONS: the VISUAL APPROACH above governs camera,
framing and composition ONLY. Every other instruction in this prompt —
location, time of day, people and their identity, poses, props,
carried state, physical rules, and the no-text rule — is binding
exactly as written earlier. If the VISUAL APPROACH appears to conflict
with any of them, the earlier instruction wins.

Edit this photograph. Apply ONLY the corrections listed below,
each in place, changing nothing else:

CORRECTIONS:
- Render the central red circle as a solid, non-glowing painted mark directly on the wood, matching the physical texture of the other circles but with a crisper crimson color.
- Remove all text, numbers, and text-like squiggles from the map in the bottom right corner and the radio devices on the wall.

PRESERVE EVERYTHING ELSE EXACTLY as in the original photograph —
every person, pose, face, garment, object, set feature, lighting,
camera and framing stay identical. No text, captions or
watermarks anywhere.

Create ONE FINAL photorealistic live-action film still of the moment below — contemporary South Korea, 2026; all people are Korean unless stated. TIME OF DAY (lock): dusk.

SHOT TEXT (authoritative, Korean): 눈을 감은 수리영의 머리가 인우의 어깨 위에 완전히 꺾여 얹혀진 정지 상태

LOCATION (lock): Inside the small fishing boat's wheelhouse beside the helm, where the fainting passenger collapses onto the boatman's shoulder. The shot takes place here — the attached PREVIOUS SHOT STILL shows this exact place.

CAMERA & FRAME (follow exactly — this is the composition authority for this still; the reference images supply identity and place, never the framing):
- CAMERA: At eye level beside both figures, the static camera views their contact in perpendicular profile: 수리영 folds in from the left and her closed-eyed head lies fully across 인우's shoulder on the right. Their joined silhouettes occupy the central band of the medium-close frame, with the changed body position—the completed collapse—held as the sole visual emphasis.
- FRAMING SCALE: medium shot
- FRAME LAYOUT: 수리영 collapsed across 인우's shoulder in the middle-left of the frame, midground; 인우 supporting 수리영 in the middle-right of the frame, midground.
- KEY BACKGROUND ELEMENTS: wheelhouse wall with circular forms (Partially obscured by both figures) — Only part of the marked face remains visible behind the figures, carrying cropped circular forms; used as Subdued rear plane preserving the claustrophobic wheelhouse enclosure around the locked two-person silhouette.
Compose the frame exactly as specified above — camera angle, subject scale and screen placement. Keep true physical scale between people and background elements: fixtures and distant objects occupy only the small screen area their real size and distance dictate; never enlarge a background object into a foreground presence.

THIS SHOT CONTINUES THE PREVIOUS SHOT: everything the attached still established — the place, its fixed features and wear, each person's clothing and state — persists. Any person in it who cannot move stays PRECISELY as photographed (body, pose, contact points, held objects); only the camera changes.

PREVIOUS STILL USAGE (follow exactly — what to take from the attached still and what to exclude): Take the same wheelhouse wall covered with red circles, enclosed dusk lighting, and the man’s established position and clothing. Exclude his earlier shocked facial pose and do not duplicate the woman as a separate standing figure; show her unconscious head resting fully on his shoulder.

IMMOBILE CHARACTER POSE — CANONICAL (identical wherever this character appears in ANY panel; on any conflict THIS POSE WINS): She is unconscious and slumped toward Inwoo with her eyes closed and her head fully bent sideways onto his shoulder; her arms and lower body hang without active support.

IMMOBILE BODIES OBEY GRAVITY: a person who is dead or unconscious
exerts NO muscular effort. Every part of their body — head, torso,
arms, hands, fingers, legs — rests fully on whatever supports it
(floor, wall, furniture, their own lap) and hangs or slumps with
gravity. NEVER show any part of an immobile person's body lifted,
raised, held up in the air, or posed as if presenting something:
an object in their grip stays clenched in a hand that itself lies
fallen on a support — the hand does not hold the object up. If the
canonical pose leaves a body part unspecified, resolve it as the
most gravity-compliant, fully-supported position.

REALIZE FIGURATIVE LANGUAGE AS A LIVE-ACTION SHOT: the Korean shot
text may describe characters metaphorically, figuratively or with
exaggeration. Photograph what a real movie camera would actually
record on a physical set — exaggerated or figurative impressions
become realistic staging choices (lighting, distance, angle,
wardrobe), not literal fantasy imagery.
EVERY CHARACTER IS A HUMAN BEING: unless the story explicitly
features non-human or virtual beings (as in science fiction or
fantasy), every character — however indirectly, vaguely or
figuratively the text describes them — IS a real human. When the
text gives no direct visual description of a person, IMAGINE one
and still show them as a concrete, fully-formed human being:
always render the human form to the maximum extent — build,
posture, face, hands, clothing — never reduce a person to a shape,
blob, solid silhouette or abstract mass.

EXPRESSIONS ARE ACTED, NEVER ANATOMICAL: when the text describes a
person's eyes, face or presence emotionally or figuratively (vacant,
hollow, dazed, burning, lifeless gaze and the like), it describes an
ACTOR'S PERFORMANCE captured by a real camera — realize it ONLY
through gaze direction, focus, eyelids, facial muscles, stillness
and posture. Every living person's eyes remain anatomically normal
human eyes with a natural iris and pupil, natural sclera and normal
proportions; NEVER whiten, blank out, cloud over, glow, enlarge or
otherwise alter eyeballs, skin or anatomy — unless the story
explicitly declares that being non-human or supernatural in form.

PROPS FACE THE RIGHT WAY: every handheld or used object must be
oriented exactly as its real-world use requires. A person reading,
watching or operating something (a phone, a photograph, a paper,
any device) has its functional side — screen, front, page — facing
THEIR OWN eyes; the camera then sees whatever side the staging
geometry implies (often its back). Show the functional side to the
camera ONLY when the shot text itself stages it toward the viewer.
Never flip, mirror or reverse an object's front and back.

CARRIED STATE (persist exactly — must match the neighbouring shots of this scene): Su-ri-young is unconscious and slumped toward Inwoo with her eyes closed and her head fully bent sideways onto his shoulder; her arms and lower body hang without active support. Her old photograph, backpack, maps, coordinate note, and phone remain aboard, and the wheelhouse walls retain their many circles.

PEOPLE: the SHOT TEXT alone decides whether any person is visible in this shot. IF a person appears, they must be one of: 수리영 (한국인 20대 초반 여성, 긴 검은 생머리, 젊은 타원형 얼굴); 인우 (한국인 20대 중반 남성, 짧은 검은 머리, 반듯한 얼굴형, 젊은 인상) — never anyone else, and never add a person the shot text does not show. IF only part of a person is in frame (a hand, arm, foot, back, silhouette), that body part belongs to the specific person the shot text names — its sex, age, build, skin and grooming must unmistakably match that person's profile above.

No text, captions, watermarks or annotations anywhere.

Create ONE FINAL photorealistic live-action film still of the moment below — contemporary South Korea, 2026; all people are Korean unless stated. TIME OF DAY (lock): dusk.

SHOT TEXT (authoritative, Korean): 눈을 감은 수리영의 머리가 인우의 어깨 위에 완전히 꺾여 얹혀진 정지 상태

LOCATION (lock): Inside the small fishing boat's wheelhouse beside the helm, where the fainting passenger collapses onto the boatman's shoulder. The shot takes place here — the attached PREVIOUS SHOT STILL shows this exact place.

CAMERA & FRAME (follow exactly — this is the composition authority for this still; the reference images supply identity and place, never the framing):
- CAMERA: At eye level beside both figures, the static camera views their contact in perpendicular profile: 수리영 folds in from the left and her closed-eyed head lies fully across 인우's shoulder on the right. Their joined silhouettes occupy the central band of the medium-close frame, with the changed body position—the completed collapse—held as the sole visual emphasis.
- FRAMING SCALE: medium shot
- FRAME LAYOUT: 수리영 collapsed across 인우's shoulder in the middle-left of the frame, midground; 인우 supporting 수리영 in the middle-right of the frame, midground.
- KEY BACKGROUND ELEMENTS: wheelhouse wall with circular forms (Partially obscured by both figures) — Only part of the marked face remains visible behind the figures, carrying cropped circular forms; used as Subdued rear plane preserving the claustrophobic wheelhouse enclosure around the locked two-person silhouette.
Compose the frame exactly as specified above — camera angle, subject scale and screen placement. Keep true physical scale between people and background elements: fixtures and distant objects occupy only the small screen area their real size and distance dictate; never enlarge a background object into a foreground presence.

THIS SHOT CONTINUES THE PREVIOUS SHOT: everything the attached still established — the place, its fixed features and wear, each person's clothing and state — persists. Any person in it who cannot move stays PRECISELY as photographed (body, pose, contact points, held objects); only the camera changes.

PREVIOUS STILL USAGE (follow exactly — what to take from the attached still and what to exclude): Take the same wheelhouse wall covered with red circles, enclosed dusk lighting, and the man’s established position and clothing. Exclude his earlier shocked facial pose and do not duplicate the woman as a separate standing figure; show her unconscious head resting fully on his shoulder.

IMMOBILE CHARACTER POSE — CANONICAL (identical wherever this character appears in ANY panel; on any conflict THIS POSE WINS): She is unconscious and slumped toward Inwoo with her eyes closed and her head fully bent sideways onto his shoulder; her arms and lower body hang without active support.

IMMOBILE BODIES OBEY GRAVITY: a person who is dead or unconscious
exerts NO muscular effort. Every part of their body — head, torso,
arms, hands, fingers, legs — rests fully on whatever supports it
(floor, wall, furniture, their own lap) and hangs or slumps with
gravity. NEVER show any part of an immobile person's body lifted,
raised, held up in the air, or posed as if presenting something:
an object in their grip stays clenched in a hand that itself lies
fallen on a support — the hand does not hold the object up. If the
canonical pose leaves a body part unspecified, resolve it as the
most gravity-compliant, fully-supported position.

REALIZE FIGURATIVE LANGUAGE AS A LIVE-ACTION SHOT: the Korean shot
text may describe characters metaphorically, figuratively or with
exaggeration. Photograph what a real movie camera would actually
record on a physical set — exaggerated or figurative impressions
become realistic staging choices (lighting, distance, angle,
wardrobe), not literal fantasy imagery.
EVERY CHARACTER IS A HUMAN BEING: unless the story explicitly
features non-human or virtual beings (as in science fiction or
fantasy), every character — however indirectly, vaguely or
figuratively the text describes them — IS a real human. When the
text gives no direct visual description of a person, IMAGINE one
and still show them as a concrete, fully-formed human being:
always render the human form to the maximum extent — build,
posture, face, hands, clothing — never reduce a person to a shape,
blob, solid silhouette or abstract mass.

EXPRESSIONS ARE ACTED, NEVER ANATOMICAL: when the text describes a
person's eyes, face or presence emotionally or figuratively (vacant,
hollow, dazed, burning, lifeless gaze and the like), it describes an
ACTOR'S PERFORMANCE captured by a real camera — realize it ONLY
through gaze direction, focus, eyelids, facial muscles, stillness
and posture. Every living person's eyes remain anatomically normal
human eyes with a natural iris and pupil, natural sclera and normal
proportions; NEVER whiten, blank out, cloud over, glow, enlarge or
otherwise alter eyeballs, skin or anatomy — unless the story
explicitly declares that being non-human or supernatural in form.

PROPS FACE THE RIGHT WAY: every handheld or used object must be
oriented exactly as its real-world use requires. A person reading,
watching or operating something (a phone, a photograph, a paper,
any device) has its functional side — screen, front, page — facing
THEIR OWN eyes; the camera then sees whatever side the staging
geometry implies (often its back). Show the functional side to the
camera ONLY when the shot text itself stages it toward the viewer.
Never flip, mirror or reverse an object's front and back.

CARRIED STATE (persist exactly — must match the neighbouring shots of this scene): Su-ri-young is unconscious and slumped toward Inwoo with her eyes closed and her head fully bent sideways onto his shoulder; her arms and lower body hang without active support. Her old photograph, backpack, maps, coordinate note, and phone remain aboard, and the wheelhouse walls retain their many circles.

PEOPLE: the SHOT TEXT alone decides whether any person is visible in this shot. IF a person appears, they must be one of: 수리영 (한국인 20대 초반 여성, 긴 검은 생머리, 젊은 타원형 얼굴); 인우 (한국인 20대 중반 남성, 짧은 검은 머리, 반듯한 얼굴형, 젊은 인상) — never anyone else, and never add a person the shot text does not show. IF only part of a person is in frame (a hand, arm, foot, back, silhouette), that body part belongs to the specific person the shot text names — its sex, age, build, skin and grooming must unmistakably match that person's profile above.

No text, captions, watermarks or annotations anywhere.

VISUAL APPROACH — this take (camera, framing, composition):
Use a restrained, intimate medium-shot treatment with a natural portrait-lens feel and shallow depth of field. Hold the perpendicular profiles and shoulder contact in crisp focus while the cropped circular forms soften into the enclosed rear plane. Let the dim dusk illumination skim gently across their faces and the bend of her neck, emphasizing her full dead weight and his quiet physical support without melodrama.

PRIORITY OF INSTRUCTIONS: the VISUAL APPROACH above governs camera,
framing and composition ONLY. Every other instruction in this prompt —
location, time of day, people and their identity, poses, props,
carried state, physical rules, and the no-text rule — is binding
exactly as written earlier. If the VISUAL APPROACH appears to conflict
with any of them, the earlier instruction wins.

Create ONE FINAL photorealistic live-action film still of the moment below — contemporary South Korea, 2026; all people are Korean unless stated. TIME OF DAY (lock): dusk.

SHOT TEXT (authoritative, Korean): 눈을 감은 수리영의 머리가 인우의 어깨 위에 완전히 꺾여 얹혀진 정지 상태

LOCATION (lock): Inside the small fishing boat's wheelhouse beside the helm, where the fainting passenger collapses onto the boatman's shoulder. The shot takes place here — the attached PREVIOUS SHOT STILL shows this exact place.

CAMERA & FRAME (follow exactly — this is the composition authority for this still; the reference images supply identity and place, never the framing):
- CAMERA: At eye level beside both figures, the static camera views their contact in perpendicular profile: 수리영 folds in from the left and her closed-eyed head lies fully across 인우's shoulder on the right. Their joined silhouettes occupy the central band of the medium-close frame, with the changed body position—the completed collapse—held as the sole visual emphasis.
- FRAMING SCALE: medium shot
- FRAME LAYOUT: 수리영 collapsed across 인우's shoulder in the middle-left of the frame, midground; 인우 supporting 수리영 in the middle-right of the frame, midground.
- KEY BACKGROUND ELEMENTS: wheelhouse wall with circular forms (Partially obscured by both figures) — Only part of the marked face remains visible behind the figures, carrying cropped circular forms; used as Subdued rear plane preserving the claustrophobic wheelhouse enclosure around the locked two-person silhouette.
Compose the frame exactly as specified above — camera angle, subject scale and screen placement. Keep true physical scale between people and background elements: fixtures and distant objects occupy only the small screen area their real size and distance dictate; never enlarge a background object into a foreground presence.

THIS SHOT CONTINUES THE PREVIOUS SHOT: everything the attached still established — the place, its fixed features and wear, each person's clothing and state — persists. Any person in it who cannot move stays PRECISELY as photographed (body, pose, contact points, held objects); only the camera changes.

PREVIOUS STILL USAGE (follow exactly — what to take from the attached still and what to exclude): Take the same wheelhouse wall covered with red circles, enclosed dusk lighting, and the man’s established position and clothing. Exclude his earlier shocked facial pose and do not duplicate the woman as a separate standing figure; show her unconscious head resting fully on his shoulder.

IMMOBILE CHARACTER POSE — CANONICAL (identical wherever this character appears in ANY panel; on any conflict THIS POSE WINS): She is unconscious and slumped toward Inwoo with her eyes closed and her head fully bent sideways onto his shoulder; her arms and lower body hang without active support.

IMMOBILE BODIES OBEY GRAVITY: a person who is dead or unconscious
exerts NO muscular effort. Every part of their body — head, torso,
arms, hands, fingers, legs — rests fully on whatever supports it
(floor, wall, furniture, their own lap) and hangs or slumps with
gravity. NEVER show any part of an immobile person's body lifted,
raised, held up in the air, or posed as if presenting something:
an object in their grip stays clenched in a hand that itself lies
fallen on a support — the hand does not hold the object up. If the
canonical pose leaves a body part unspecified, resolve it as the
most gravity-compliant, fully-supported position.

REALIZE FIGURATIVE LANGUAGE AS A LIVE-ACTION SHOT: the Korean shot
text may describe characters metaphorically, figuratively or with
exaggeration. Photograph what a real movie camera would actually
record on a physical set — exaggerated or figurative impressions
become realistic staging choices (lighting, distance, angle,
wardrobe), not literal fantasy imagery.
EVERY CHARACTER IS A HUMAN BEING: unless the story explicitly
features non-human or virtual beings (as in science fiction or
fantasy), every character — however indirectly, vaguely or
figuratively the text describes them — IS a real human. When the
text gives no direct visual description of a person, IMAGINE one
and still show them as a concrete, fully-formed human being:
always render the human form to the maximum extent — build,
posture, face, hands, clothing — never reduce a person to a shape,
blob, solid silhouette or abstract mass.

EXPRESSIONS ARE ACTED, NEVER ANATOMICAL: when the text describes a
person's eyes, face or presence emotionally or figuratively (vacant,
hollow, dazed, burning, lifeless gaze and the like), it describes an
ACTOR'S PERFORMANCE captured by a real camera — realize it ONLY
through gaze direction, focus, eyelids, facial muscles, stillness
and posture. Every living person's eyes remain anatomically normal
human eyes with a natural iris and pupil, natural sclera and normal
proportions; NEVER whiten, blank out, cloud over, glow, enlarge or
otherwise alter eyeballs, skin or anatomy — unless the story
explicitly declares that being non-human or supernatural in form.

PROPS FACE THE RIGHT WAY: every handheld or used object must be
oriented exactly as its real-world use requires. A person reading,
watching or operating something (a phone, a photograph, a paper,
any device) has its functional side — screen, front, page — facing
THEIR OWN eyes; the camera then sees whatever side the staging
geometry implies (often its back). Show the functional side to the
camera ONLY when the shot text itself stages it toward the viewer.
Never flip, mirror or reverse an object's front and back.

CARRIED STATE (persist exactly — must match the neighbouring shots of this scene): Su-ri-young is unconscious and slumped toward Inwoo with her eyes closed and her head fully bent sideways onto his shoulder; her arms and lower body hang without active support. Her old photograph, backpack, maps, coordinate note, and phone remain aboard, and the wheelhouse walls retain their many circles.

PEOPLE: the SHOT TEXT alone decides whether any person is visible in this shot. IF a person appears, they must be one of: 수리영 (한국인 20대 초반 여성, 긴 검은 생머리, 젊은 타원형 얼굴); 인우 (한국인 20대 중반 남성, 짧은 검은 머리, 반듯한 얼굴형, 젊은 인상) — never anyone else, and never add a person the shot text does not show. IF only part of a person is in frame (a hand, arm, foot, back, silhouette), that body part belongs to the specific person the shot text names — its sex, age, build, skin and grooming must unmistakably match that person's profile above.

No text, captions, watermarks or annotations anywhere.

VISUAL APPROACH — this take (camera, framing, composition):
Photograph the same eye-level medium profile as a compressed, nearly planar composition with deeper focus and firmer edge definition. Keep both joined figures and the partially visible circular wall markings legible, using the cramped rear plane to visually pin their locked silhouette in place. Favor low-key, subdued dusk contrast with broad shadow masses, making the completed collapse feel stark, static, and claustrophobic.

PRIORITY OF INSTRUCTIONS: the VISUAL APPROACH above governs camera,
framing and composition ONLY. Every other instruction in this prompt —
location, time of day, people and their identity, poses, props,
carried state, physical rules, and the no-text rule — is binding
exactly as written earlier. If the VISUAL APPROACH appears to conflict
with any of them, the earlier instruction wins.

Edit this photograph. Apply ONLY the corrections listed below,
each in place, changing nothing else:

CORRECTIONS:
- Adjust Su-ri-young's posture so her neck is sharply bent and her head rests fully on top of Inwoo's shoulder, not against his back.
- Add the olive green cap from the character reference to Su-ri-young's head.
- Add the red scratches and marks to Su-ri-young's right cheek and chin exactly as they appear in her character reference.

PRESERVE EVERYTHING ELSE EXACTLY as in the original photograph —
every person, pose, face, garment, object, set feature, lighting,
camera and framing stay identical. No text, captions or
watermarks anywhere.

Create ONE FINAL photorealistic live-action film still of the moment below — contemporary South Korea, 2026; all people are Korean unless stated. TIME OF DAY (lock): day.

SHOT TEXT (authoritative, Korean): 햇빛이 비치는 인천 변두리의 낡은 빌라 옥탑방 전경 부감 샷

LOCATION (lock): Above an aging villa on the city outskirts, overlooking the rooftop structure and its sunlit exterior. The shot takes place here — the attached LOCATION PHOTOGRAPH shows the exact spot.

STRUCTURE LOOK AUTHORITY: the attached STRUCTURE LOOK photograph is the identity of the fixed structure at this location — wherever that structure appears in the frame, its shape, proportions, openings, materials and colors are LOCKED to it. The LOCATION PHOTOGRAPH remains the authority for this shot's sub-space, surroundings, time of day and lighting. If the two conflict on the structure itself, the STRUCTURE LOOK photo wins; for everything else, the LOCATION PHOTOGRAPH wins.

CAMERA & FRAME (follow exactly — this is the composition authority for this still; the reference images supply identity and place, never the framing):
- CAMERA: From the closer overhead position, the crane completes its forward descent on the established aerial axis, angling steeply down so the old villa rooftop room sits near center while adjacent outskirts remain visible around all sides. The rooftop room receives emphasis through placement rather than excessive scale, occupying well under half the frame and preserving the wider neighborhood context.
- FRAMING SCALE: wide shot
- FRAME LAYOUT: old villa rooftop room in the middle-center of the frame, midground; Incheon outskirts in the upper-center of the frame, background.
- KEY BACKGROUND ELEMENTS: 오래된 빌라 옥탑방 (Old and sunlit) — Its roof and surrounding rooftop-facing sides are visible from above; used as Central architectural focal point at the end of the aerial approach; 인천 변두리 건물들 (Daytime exterior) — Roofs and upper building faces are visible from the overhead angle; used as Surrounding spatial context retained around the central building.
Compose the frame exactly as specified above — camera angle, subject scale and screen placement. Keep true physical scale between people and background elements: fixtures and distant objects occupy only the small screen area their real size and distance dictate; never enlarge a background object into a foreground presence.

REALIZE FIGURATIVE LANGUAGE AS A LIVE-ACTION SHOT: the Korean shot
text may describe characters metaphorically, figuratively or with
exaggeration. Photograph what a real movie camera would actually
record on a physical set — exaggerated or figurative impressions
become realistic staging choices (lighting, distance, angle,
wardrobe), not literal fantasy imagery.
EVERY CHARACTER IS A HUMAN BEING: unless the story explicitly
features non-human or virtual beings (as in science fiction or
fantasy), every character — however indirectly, vaguely or
figuratively the text describes them — IS a real human. When the
text gives no direct visual description of a person, IMAGINE one
and still show them as a concrete, fully-formed human being:
always render the human form to the maximum extent — build,
posture, face, hands, clothing — never reduce a person to a shape,
blob, solid silhouette or abstract mass.

EXPRESSIONS ARE ACTED, NEVER ANATOMICAL: when the text describes a
person's eyes, face or presence emotionally or figuratively (vacant,
hollow, dazed, burning, lifeless gaze and the like), it describes an
ACTOR'S PERFORMANCE captured by a real camera — realize it ONLY
through gaze direction, focus, eyelids, facial muscles, stillness
and posture. Every living person's eyes remain anatomically normal
human eyes with a natural iris and pupil, natural sclera and normal
proportions; NEVER whiten, blank out, cloud over, glow, enlarge or
otherwise alter eyeballs, skin or anatomy — unless the story
explicitly declares that being non-human or supernatural in form.

PROPS FACE THE RIGHT WAY: every handheld or used object must be
oriented exactly as its real-world use requires. A person reading,
watching or operating something (a phone, a photograph, a paper,
any device) has its functional side — screen, front, page — facing
THEIR OWN eyes; the camera then sees whatever side the staging
geometry implies (often its back). Show the functional side to the
camera ONLY when the shot text itself stages it toward the viewer.
Never flip, mirror or reverse an object's front and back.

NO PEOPLE IN THIS SHOT: the shot text shows only the place and its state — no living person or any body part appears in frame, unless the shot text itself explicitly says so.

No text, captions, watermarks or annotations anywhere.

Create ONE FINAL photorealistic live-action film still of the moment below — contemporary South Korea, 2026; all people are Korean unless stated. TIME OF DAY (lock): day.

SHOT TEXT (authoritative, Korean): 햇빛이 비치는 인천 변두리의 낡은 빌라 옥탑방 전경 부감 샷

LOCATION (lock): Above an aging villa on the city outskirts, overlooking the rooftop structure and its sunlit exterior. The shot takes place here — the attached LOCATION PHOTOGRAPH shows the exact spot.

STRUCTURE LOOK AUTHORITY: the attached STRUCTURE LOOK photograph is the identity of the fixed structure at this location — wherever that structure appears in the frame, its shape, proportions, openings, materials and colors are LOCKED to it. The LOCATION PHOTOGRAPH remains the authority for this shot's sub-space, surroundings, time of day and lighting. If the two conflict on the structure itself, the STRUCTURE LOOK photo wins; for everything else, the LOCATION PHOTOGRAPH wins.

CAMERA & FRAME (follow exactly — this is the composition authority for this still; the reference images supply identity and place, never the framing):
- CAMERA: From the closer overhead position, the crane completes its forward descent on the established aerial axis, angling steeply down so the old villa rooftop room sits near center while adjacent outskirts remain visible around all sides. The rooftop room receives emphasis through placement rather than excessive scale, occupying well under half the frame and preserving the wider neighborhood context.
- FRAMING SCALE: wide shot
- FRAME LAYOUT: old villa rooftop room in the middle-center of the frame, midground; Incheon outskirts in the upper-center of the frame, background.
- KEY BACKGROUND ELEMENTS: 오래된 빌라 옥탑방 (Old and sunlit) — Its roof and surrounding rooftop-facing sides are visible from above; used as Central architectural focal point at the end of the aerial approach; 인천 변두리 건물들 (Daytime exterior) — Roofs and upper building faces are visible from the overhead angle; used as Surrounding spatial context retained around the central building.
Compose the frame exactly as specified above — camera angle, subject scale and screen placement. Keep true physical scale between people and background elements: fixtures and distant objects occupy only the small screen area their real size and distance dictate; never enlarge a background object into a foreground presence.

REALIZE FIGURATIVE LANGUAGE AS A LIVE-ACTION SHOT: the Korean shot
text may describe characters metaphorically, figuratively or with
exaggeration. Photograph what a real movie camera would actually
record on a physical set — exaggerated or figurative impressions
become realistic staging choices (lighting, distance, angle,
wardrobe), not literal fantasy imagery.
EVERY CHARACTER IS A HUMAN BEING: unless the story explicitly
features non-human or virtual beings (as in science fiction or
fantasy), every character — however indirectly, vaguely or
figuratively the text describes them — IS a real human. When the
text gives no direct visual description of a person, IMAGINE one
and still show them as a concrete, fully-formed human being:
always render the human form to the maximum extent — build,
posture, face, hands, clothing — never reduce a person to a shape,
blob, solid silhouette or abstract mass.

EXPRESSIONS ARE ACTED, NEVER ANATOMICAL: when the text describes a
person's eyes, face or presence emotionally or figuratively (vacant,
hollow, dazed, burning, lifeless gaze and the like), it describes an
ACTOR'S PERFORMANCE captured by a real camera — realize it ONLY
through gaze direction, focus, eyelids, facial muscles, stillness
and posture. Every living person's eyes remain anatomically normal
human eyes with a natural iris and pupil, natural sclera and normal
proportions; NEVER whiten, blank out, cloud over, glow, enlarge or
otherwise alter eyeballs, skin or anatomy — unless the story
explicitly declares that being non-human or supernatural in form.

PROPS FACE THE RIGHT WAY: every handheld or used object must be
oriented exactly as its real-world use requires. A person reading,
watching or operating something (a phone, a photograph, a paper,
any device) has its functional side — screen, front, page — facing
THEIR OWN eyes; the camera then sees whatever side the staging
geometry implies (often its back). Show the functional side to the
camera ONLY when the shot text itself stages it toward the viewer.
Never flip, mirror or reverse an object's front and back.

NO PEOPLE IN THIS SHOT: the shot text shows only the place and its state — no living person or any body part appears in frame, unless the shot text itself explicitly says so.

No text, captions, watermarks or annotations anywhere.

VISUAL APPROACH — this take (camera, framing, composition):
Use a rectilinear wide-lens feel with strong depth of field, rendering the overhead geometry and surrounding spatial relationships with calm, topographic clarity. Preserve the reference sunlight direction and let its crisp tonal separation reveal aged surfaces without exaggerating contrast. Keep the composition stable and observational, with clean verticals and restrained cinematic color.

PRIORITY OF INSTRUCTIONS: the VISUAL APPROACH above governs camera,
framing and composition ONLY. Every other instruction in this prompt —
location, time of day, people and their identity, poses, props,
carried state, physical rules, and the no-text rule — is binding
exactly as written earlier. If the VISUAL APPROACH appears to conflict
with any of them, the earlier instruction wins.

Create ONE FINAL photorealistic live-action film still of the moment below — contemporary South Korea, 2026; all people are Korean unless stated. TIME OF DAY (lock): day.

SHOT TEXT (authoritative, Korean): 햇빛이 비치는 인천 변두리의 낡은 빌라 옥탑방 전경 부감 샷

LOCATION (lock): Above an aging villa on the city outskirts, overlooking the rooftop structure and its sunlit exterior. The shot takes place here — the attached LOCATION PHOTOGRAPH shows the exact spot.

STRUCTURE LOOK AUTHORITY: the attached STRUCTURE LOOK photograph is the identity of the fixed structure at this location — wherever that structure appears in the frame, its shape, proportions, openings, materials and colors are LOCKED to it. The LOCATION PHOTOGRAPH remains the authority for this shot's sub-space, surroundings, time of day and lighting. If the two conflict on the structure itself, the STRUCTURE LOOK photo wins; for everything else, the LOCATION PHOTOGRAPH wins.

CAMERA & FRAME (follow exactly — this is the composition authority for this still; the reference images supply identity and place, never the framing):
- CAMERA: From the closer overhead position, the crane completes its forward descent on the established aerial axis, angling steeply down so the old villa rooftop room sits near center while adjacent outskirts remain visible around all sides. The rooftop room receives emphasis through placement rather than excessive scale, occupying well under half the frame and preserving the wider neighborhood context.
- FRAMING SCALE: wide shot
- FRAME LAYOUT: old villa rooftop room in the middle-center of the frame, midground; Incheon outskirts in the upper-center of the frame, background.
- KEY BACKGROUND ELEMENTS: 오래된 빌라 옥탑방 (Old and sunlit) — Its roof and surrounding rooftop-facing sides are visible from above; used as Central architectural focal point at the end of the aerial approach; 인천 변두리 건물들 (Daytime exterior) — Roofs and upper building faces are visible from the overhead angle; used as Surrounding spatial context retained around the central building.
Compose the frame exactly as specified above — camera angle, subject scale and screen placement. Keep true physical scale between people and background elements: fixtures and distant objects occupy only the small screen area their real size and distance dictate; never enlarge a background object into a foreground presence.

REALIZE FIGURATIVE LANGUAGE AS A LIVE-ACTION SHOT: the Korean shot
text may describe characters metaphorically, figuratively or with
exaggeration. Photograph what a real movie camera would actually
record on a physical set — exaggerated or figurative impressions
become realistic staging choices (lighting, distance, angle,
wardrobe), not literal fantasy imagery.
EVERY CHARACTER IS A HUMAN BEING: unless the story explicitly
features non-human or virtual beings (as in science fiction or
fantasy), every character — however indirectly, vaguely or
figuratively the text describes them — IS a real human. When the
text gives no direct visual description of a person, IMAGINE one
and still show them as a concrete, fully-formed human being:
always render the human form to the maximum extent — build,
posture, face, hands, clothing — never reduce a person to a shape,
blob, solid silhouette or abstract mass.

EXPRESSIONS ARE ACTED, NEVER ANATOMICAL: when the text describes a
person's eyes, face or presence emotionally or figuratively (vacant,
hollow, dazed, burning, lifeless gaze and the like), it describes an
ACTOR'S PERFORMANCE captured by a real camera — realize it ONLY
through gaze direction, focus, eyelids, facial muscles, stillness
and posture. Every living person's eyes remain anatomically normal
human eyes with a natural iris and pupil, natural sclera and normal
proportions; NEVER whiten, blank out, cloud over, glow, enlarge or
otherwise alter eyeballs, skin or anatomy — unless the story
explicitly declares that being non-human or supernatural in form.

PROPS FACE THE RIGHT WAY: every handheld or used object must be
oriented exactly as its real-world use requires. A person reading,
watching or operating something (a phone, a photograph, a paper,
any device) has its functional side — screen, front, page — facing
THEIR OWN eyes; the camera then sees whatever side the staging
geometry implies (often its back). Show the functional side to the
camera ONLY when the shot text itself stages it toward the viewer.
Never flip, mirror or reverse an object's front and back.

NO PEOPLE IN THIS SHOT: the shot text shows only the place and its state — no living person or any body part appears in frame, unless the shot text itself explicitly says so.

No text, captions, watermarks or annotations anywhere.

VISUAL APPROACH — this take (camera, framing, composition):
Use a subtly compressed lens feel from the locked overhead axis, visually layering the central architecture against the surrounding outskirts while retaining context on every side. Hold the focal emphasis on the middle-center and allow detail toward the far edges to soften gently through natural aerial depth rather than blur. Shape the reference daylight with a warmer highlight roll-off and subdued peripheral contrast, giving the completed descent a quiet sense of arrival.

PRIORITY OF INSTRUCTIONS: the VISUAL APPROACH above governs camera,
framing and composition ONLY. Every other instruction in this prompt —
location, time of day, people and their identity, poses, props,
carried state, physical rules, and the no-text rule — is binding
exactly as written earlier. If the VISUAL APPROACH appears to conflict
with any of them, the earlier instruction wins.

Edit this photograph. Apply ONLY the corrections listed below,
each in place, changing nothing else:

CORRECTIONS:
- Move the external staircase from the front facade to run along the right side wall of the building.
- Place the grey door on the same front-facing wall as the window on the white rooftop room.

PRESERVE EVERYTHING ELSE EXACTLY as in the original photograph —
every person, pose, face, garment, object, set feature, lighting,
camera and framing stay identical. No text, captions or
watermarks anywhere.

Create ONE FINAL photorealistic live-action film still of the moment below — contemporary South Korea, 2026; all people are Korean unless stated. TIME OF DAY (lock): night.

SHOT TEXT (authoritative, Korean): 창문에 비친 그림자로, 민숙을 습격한 낯선 남자의 실루엣이 강민숙의 실루엣 어깨 쪽으로 입을 벌리며 덮치려는 찰나의 측면 클로즈업

LOCATION (lock): Outside the rooftop apartment at night, facing a window that silhouettes the assault occurring inside. The shot takes place here — the attached LOCATION PHOTOGRAPH shows the exact spot.

STRUCTURE LOOK AUTHORITY: the attached STRUCTURE LOOK photograph is the identity of the fixed structure at this location — wherever that structure appears in the frame, its shape, proportions, openings, materials and colors are LOCKED to it. The LOCATION PHOTOGRAPH remains the authority for this shot's sub-space, surroundings, time of day and lighting. If the two conflict on the structure itself, the STRUCTURE LOOK photo wins; for everything else, the LOCATION PHOTOGRAPH wins.

CAMERA & FRAME (follow exactly — this is the composition authority for this still; the reference images supply identity and place, never the framing):
- CAMERA: Outside the window at shoulder height, the camera sits close and perpendicular to the pane, holding a static side close-up of the reflected silhouettes with the reflection visibly left-right reversed. 강민숙’s shoulder occupies the middle-right while 민숙을 습격한 낯선 남자 crowds in from the middle-left, his open mouth arrested just before contact.
- FRAMING SCALE: close-up
- FRAME LAYOUT: 민숙을 습격한 낯선 남자 reflected silhouette in the middle-left of the frame, midground, moves toward 강민숙's shoulder; 강민숙 reflected silhouette in the middle-right of the frame, midground.
- KEY BACKGROUND ELEMENTS: 옥탑방 창문 (Reflecting the attack silhouettes) — The pane reflects the two overlapping side silhouettes in left-right reversal, with the attacker closing on 강민숙’s shoulder; used as Reflective plane containing the entire attack image and withholding facial detail.
Compose the frame exactly as specified above — camera angle, subject scale and screen placement. Keep true physical scale between people and background elements: fixtures and distant objects occupy only the small screen area their real size and distance dictate; never enlarge a background object into a foreground presence.

REALIZE FIGURATIVE LANGUAGE AS A LIVE-ACTION SHOT: the Korean shot
text may describe characters metaphorically, figuratively or with
exaggeration. Photograph what a real movie camera would actually
record on a physical set — exaggerated or figurative impressions
become realistic staging choices (lighting, distance, angle,
wardrobe), not literal fantasy imagery.
EVERY CHARACTER IS A HUMAN BEING: unless the story explicitly
features non-human or virtual beings (as in science fiction or
fantasy), every character — however indirectly, vaguely or
figuratively the text describes them — IS a real human. When the
text gives no direct visual description of a person, IMAGINE one
and still show them as a concrete, fully-formed human being:
always render the human form to the maximum extent — build,
posture, face, hands, clothing — never reduce a person to a shape,
blob, solid silhouette or abstract mass.

EXPRESSIONS ARE ACTED, NEVER ANATOMICAL: when the text describes a
person's eyes, face or presence emotionally or figuratively (vacant,
hollow, dazed, burning, lifeless gaze and the like), it describes an
ACTOR'S PERFORMANCE captured by a real camera — realize it ONLY
through gaze direction, focus, eyelids, facial muscles, stillness
and posture. Every living person's eyes remain anatomically normal
human eyes with a natural iris and pupil, natural sclera and normal
proportions; NEVER whiten, blank out, cloud over, glow, enlarge or
otherwise alter eyeballs, skin or anatomy — unless the story
explicitly declares that being non-human or supernatural in form.

PROPS FACE THE RIGHT WAY: every handheld or used object must be
oriented exactly as its real-world use requires. A person reading,
watching or operating something (a phone, a photograph, a paper,
any device) has its functional side — screen, front, page — facing
THEIR OWN eyes; the camera then sees whatever side the staging
geometry implies (often its back). Show the functional side to the
camera ONLY when the shot text itself stages it toward the viewer.
Never flip, mirror or reverse an object's front and back.

CARRIED STATE (persist exactly — must match the neighbouring shots of this scene): The intruder has left the iron entrance door slightly open after entering. Min-suk still bears the small circular mark on her wrist as she is attacked.

NO PEOPLE IN THIS SHOT: the shot text shows only the place and its state — no living person or any body part appears in frame, unless the shot text itself explicitly says so.

No text, captions, watermarks or annotations anywhere.

Create ONE FINAL photorealistic live-action film still of the moment below — contemporary South Korea, 2026; all people are Korean unless stated. TIME OF DAY (lock): night.

SHOT TEXT (authoritative, Korean): 창문에 비친 그림자로, 민숙을 습격한 낯선 남자의 실루엣이 강민숙의 실루엣 어깨 쪽으로 입을 벌리며 덮치려는 찰나의 측면 클로즈업

LOCATION (lock): Outside the rooftop apartment at night, facing a window that silhouettes the assault occurring inside. The shot takes place here — the attached LOCATION PHOTOGRAPH shows the exact spot.

STRUCTURE LOOK AUTHORITY: the attached STRUCTURE LOOK photograph is the identity of the fixed structure at this location — wherever that structure appears in the frame, its shape, proportions, openings, materials and colors are LOCKED to it. The LOCATION PHOTOGRAPH remains the authority for this shot's sub-space, surroundings, time of day and lighting. If the two conflict on the structure itself, the STRUCTURE LOOK photo wins; for everything else, the LOCATION PHOTOGRAPH wins.

CAMERA & FRAME (follow exactly — this is the composition authority for this still; the reference images supply identity and place, never the framing):
- CAMERA: Outside the window at shoulder height, the camera sits close and perpendicular to the pane, holding a static side close-up of the reflected silhouettes with the reflection visibly left-right reversed. 강민숙’s shoulder occupies the middle-right while 민숙을 습격한 낯선 남자 crowds in from the middle-left, his open mouth arrested just before contact.
- FRAMING SCALE: close-up
- FRAME LAYOUT: 민숙을 습격한 낯선 남자 reflected silhouette in the middle-left of the frame, midground, moves toward 강민숙's shoulder; 강민숙 reflected silhouette in the middle-right of the frame, midground.
- KEY BACKGROUND ELEMENTS: 옥탑방 창문 (Reflecting the attack silhouettes) — The pane reflects the two overlapping side silhouettes in left-right reversal, with the attacker closing on 강민숙’s shoulder; used as Reflective plane containing the entire attack image and withholding facial detail.
Compose the frame exactly as specified above — camera angle, subject scale and screen placement. Keep true physical scale between people and background elements: fixtures and distant objects occupy only the small screen area their real size and distance dictate; never enlarge a background object into a foreground presence.

REALIZE FIGURATIVE LANGUAGE AS A LIVE-ACTION SHOT: the Korean shot
text may describe characters metaphorically, figuratively or with
exaggeration. Photograph what a real movie camera would actually
record on a physical set — exaggerated or figurative impressions
become realistic staging choices (lighting, distance, angle,
wardrobe), not literal fantasy imagery.
EVERY CHARACTER IS A HUMAN BEING: unless the story explicitly
features non-human or virtual beings (as in science fiction or
fantasy), every character — however indirectly, vaguely or
figuratively the text describes them — IS a real human. When the
text gives no direct visual description of a person, IMAGINE one
and still show them as a concrete, fully-formed human being:
always render the human form to the maximum extent — build,
posture, face, hands, clothing — never reduce a person to a shape,
blob, solid silhouette or abstract mass.

EXPRESSIONS ARE ACTED, NEVER ANATOMICAL: when the text describes a
person's eyes, face or presence emotionally or figuratively (vacant,
hollow, dazed, burning, lifeless gaze and the like), it describes an
ACTOR'S PERFORMANCE captured by a real camera — realize it ONLY
through gaze direction, focus, eyelids, facial muscles, stillness
and posture. Every living person's eyes remain anatomically normal
human eyes with a natural iris and pupil, natural sclera and normal
proportions; NEVER whiten, blank out, cloud over, glow, enlarge or
otherwise alter eyeballs, skin or anatomy — unless the story
explicitly declares that being non-human or supernatural in form.

PROPS FACE THE RIGHT WAY: every handheld or used object must be
oriented exactly as its real-world use requires. A person reading,
watching or operating something (a phone, a photograph, a paper,
any device) has its functional side — screen, front, page — facing
THEIR OWN eyes; the camera then sees whatever side the staging
geometry implies (often its back). Show the functional side to the
camera ONLY when the shot text itself stages it toward the viewer.
Never flip, mirror or reverse an object's front and back.

CARRIED STATE (persist exactly — must match the neighbouring shots of this scene): The intruder has left the iron entrance door slightly open after entering. Min-suk still bears the small circular mark on her wrist as she is attacked.

NO PEOPLE IN THIS SHOT: the shot text shows only the place and its state — no living person or any body part appears in frame, unless the shot text itself explicitly says so.

No text, captions, watermarks or annotations anywhere.

VISUAL APPROACH — this take (camera, framing, composition):
Use a compressed, intimate lens feel with very shallow depth of field, holding the focus precisely on the reflected profiles and the sliver of space between the open mouth and the shoulder. Let the night-lit pane fall softly away around them, shaping the overlapping human silhouettes with crisp edge separation while preserving only the limited facial information naturally visible in the reflection. Emphasize suspended motion and suffocating proximity.

PRIORITY OF INSTRUCTIONS: the VISUAL APPROACH above governs camera,
framing and composition ONLY. Every other instruction in this prompt —
location, time of day, people and their identity, poses, props,
carried state, physical rules, and the no-text rule — is binding
exactly as written earlier. If the VISUAL APPROACH appears to conflict
with any of them, the earlier instruction wins.

Create ONE FINAL photorealistic live-action film still of the moment below — contemporary South Korea, 2026; all people are Korean unless stated. TIME OF DAY (lock): night.

SHOT TEXT (authoritative, Korean): 창문에 비친 그림자로, 민숙을 습격한 낯선 남자의 실루엣이 강민숙의 실루엣 어깨 쪽으로 입을 벌리며 덮치려는 찰나의 측면 클로즈업

LOCATION (lock): Outside the rooftop apartment at night, facing a window that silhouettes the assault occurring inside. The shot takes place here — the attached LOCATION PHOTOGRAPH shows the exact spot.

STRUCTURE LOOK AUTHORITY: the attached STRUCTURE LOOK photograph is the identity of the fixed structure at this location — wherever that structure appears in the frame, its shape, proportions, openings, materials and colors are LOCKED to it. The LOCATION PHOTOGRAPH remains the authority for this shot's sub-space, surroundings, time of day and lighting. If the two conflict on the structure itself, the STRUCTURE LOOK photo wins; for everything else, the LOCATION PHOTOGRAPH wins.

CAMERA & FRAME (follow exactly — this is the composition authority for this still; the reference images supply identity and place, never the framing):
- CAMERA: Outside the window at shoulder height, the camera sits close and perpendicular to the pane, holding a static side close-up of the reflected silhouettes with the reflection visibly left-right reversed. 강민숙’s shoulder occupies the middle-right while 민숙을 습격한 낯선 남자 crowds in from the middle-left, his open mouth arrested just before contact.
- FRAMING SCALE: close-up
- FRAME LAYOUT: 민숙을 습격한 낯선 남자 reflected silhouette in the middle-left of the frame, midground, moves toward 강민숙's shoulder; 강민숙 reflected silhouette in the middle-right of the frame, midground.
- KEY BACKGROUND ELEMENTS: 옥탑방 창문 (Reflecting the attack silhouettes) — The pane reflects the two overlapping side silhouettes in left-right reversal, with the attacker closing on 강민숙’s shoulder; used as Reflective plane containing the entire attack image and withholding facial detail.
Compose the frame exactly as specified above — camera angle, subject scale and screen placement. Keep true physical scale between people and background elements: fixtures and distant objects occupy only the small screen area their real size and distance dictate; never enlarge a background object into a foreground presence.

REALIZE FIGURATIVE LANGUAGE AS A LIVE-ACTION SHOT: the Korean shot
text may describe characters metaphorically, figuratively or with
exaggeration. Photograph what a real movie camera would actually
record on a physical set — exaggerated or figurative impressions
become realistic staging choices (lighting, distance, angle,
wardrobe), not literal fantasy imagery.
EVERY CHARACTER IS A HUMAN BEING: unless the story explicitly
features non-human or virtual beings (as in science fiction or
fantasy), every character — however indirectly, vaguely or
figuratively the text describes them — IS a real human. When the
text gives no direct visual description of a person, IMAGINE one
and still show them as a concrete, fully-formed human being:
always render the human form to the maximum extent — build,
posture, face, hands, clothing — never reduce a person to a shape,
blob, solid silhouette or abstract mass.

EXPRESSIONS ARE ACTED, NEVER ANATOMICAL: when the text describes a
person's eyes, face or presence emotionally or figuratively (vacant,
hollow, dazed, burning, lifeless gaze and the like), it describes an
ACTOR'S PERFORMANCE captured by a real camera — realize it ONLY
through gaze direction, focus, eyelids, facial muscles, stillness
and posture. Every living person's eyes remain anatomically normal
human eyes with a natural iris and pupil, natural sclera and normal
proportions; NEVER whiten, blank out, cloud over, glow, enlarge or
otherwise alter eyeballs, skin or anatomy — unless the story
explicitly declares that being non-human or supernatural in form.

PROPS FACE THE RIGHT WAY: every handheld or used object must be
oriented exactly as its real-world use requires. A person reading,
watching or operating something (a phone, a photograph, a paper,
any device) has its functional side — screen, front, page — facing
THEIR OWN eyes; the camera then sees whatever side the staging
geometry implies (often its back). Show the functional side to the
camera ONLY when the shot text itself stages it toward the viewer.
Never flip, mirror or reverse an object's front and back.

CARRIED STATE (persist exactly — must match the neighbouring shots of this scene): The intruder has left the iron entrance door slightly open after entering. Min-suk still bears the small circular mark on her wrist as she is attacked.

NO PEOPLE IN THIS SHOT: the shot text shows only the place and its state — no living person or any body part appears in frame, unless the shot text itself explicitly says so.

No text, captions, watermarks or annotations anywhere.

VISUAL APPROACH — this take (camera, framing, composition):
Photograph the close-up with a more observational, flatter lens feel and comparatively deep focus, keeping the window surface and the full reflected overlap legible at once. Use restrained night contrast so the pane remains a tangible barrier while the side profiles read as layered human reflections, with the attacker pressing visually into Min-suk’s shoulder line. Favor rigid geometry and stillness, making the frozen instant feel clinical and inescapable.

PRIORITY OF INSTRUCTIONS: the VISUAL APPROACH above governs camera,
framing and composition ONLY. Every other instruction in this prompt —
location, time of day, people and their identity, poses, props,
carried state, physical rules, and the no-text rule — is binding
exactly as written earlier. If the VISUAL APPROACH appears to conflict
with any of them, the earlier instruction wins.

Edit this photograph. Apply ONLY the corrections listed below,
each in place, changing nothing else:

CORRECTIONS:
- Render the window glass as a dark reflective plane showing only the solid silhouettes of the figures on its surface, completely obscuring the interior and facial details.
- Change the window structure to exactly match the small, two-pane sliding window set into a solid wall from the reference images.
- Lower the attacker's head so his open mouth is aimed downward and positioned directly next to the woman's shoulder.

PRESERVE EVERYTHING ELSE EXACTLY as in the original photograph —
every person, pose, face, garment, object, set feature, lighting,
camera and framing stay identical. No text, captions or
watermarks anywhere.

Create ONE FINAL photorealistic live-action film still of the moment below — contemporary South Korea, 2026; all people are Korean unless stated. TIME OF DAY (lock): night.

SHOT TEXT (authoritative, Korean): 죽은 강민숙의 굳어버린 손안에 구겨진 종이 사진이 쥐어져 있는 클로즈업

LOCATION (lock): On the floor behind the curtain in the daughter's bedroom, beside the dead mother's body in the dim nighttime apartment. The shot takes place here — the attached PREVIOUS SHOT STILL shows this exact place.

CAMERA & FRAME (follow exactly — this is the composition authority for this still; the reference images supply identity and place, never the framing):
- CAMERA: At hand height, a steep but not overhead downward angle isolates 강민숙’s rigid fingers and the crumpled photograph while retaining a narrow strip of her forearm and the nearby floor for scale. The hand lies across the lower-center and the photograph remains below forty percent of the frame, its creased image side caught between the closed fingers.
- FRAMING SCALE: insert close-up on a detail
- KEY BACKGROUND ELEMENTS: 구겨진 종이 사진 (Crumpled and gripped inside the hand) — The camera sees the crumpled image-bearing face, partially concealed by fingers and folds; used as Central clue held within the rigid hand; 바닥 (No additional state specified in this framing); used as Scale reference at the edge of the isolated detail.
Compose the frame exactly as specified above — camera angle, subject scale and screen placement. Keep true physical scale between people and background elements: fixtures and distant objects occupy only the small screen area their real size and distance dictate; never enlarge a background object into a foreground presence.

THIS SHOT CONTINUES THE PREVIOUS SHOT: everything the attached still established — the place, its fixed features and wear, each person's clothing and state — persists. Any person in it who cannot move stays PRECISELY as photographed (body, pose, contact points, held objects); only the camera changes.

REALIZE FIGURATIVE LANGUAGE AS A LIVE-ACTION SHOT: the Korean shot
text may describe characters metaphorically, figuratively or with
exaggeration. Photograph what a real movie camera would actually
record on a physical set — exaggerated or figurative impressions
become realistic staging choices (lighting, distance, angle,
wardrobe), not literal fantasy imagery.
EVERY CHARACTER IS A HUMAN BEING: unless the story explicitly
features non-human or virtual beings (as in science fiction or
fantasy), every character — however indirectly, vaguely or
figuratively the text describes them — IS a real human. When the
text gives no direct visual description of a person, IMAGINE one
and still show them as a concrete, fully-formed human being:
always render the human form to the maximum extent — build,
posture, face, hands, clothing — never reduce a person to a shape,
blob, solid silhouette or abstract mass.

EXPRESSIONS ARE ACTED, NEVER ANATOMICAL: when the text describes a
person's eyes, face or presence emotionally or figuratively (vacant,
hollow, dazed, burning, lifeless gaze and the like), it describes an
ACTOR'S PERFORMANCE captured by a real camera — realize it ONLY
through gaze direction, focus, eyelids, facial muscles, stillness
and posture. Every living person's eyes remain anatomically normal
human eyes with a natural iris and pupil, natural sclera and normal
proportions; NEVER whiten, blank out, cloud over, glow, enlarge or
otherwise alter eyeballs, skin or anatomy — unless the story
explicitly declares that being non-human or supernatural in form.

PROPS FACE THE RIGHT WAY: every handheld or used object must be
oriented exactly as its real-world use requires. A person reading,
watching or operating something (a phone, a photograph, a paper,
any device) has its functional side — screen, front, page — facing
THEIR OWN eyes; the camera then sees whatever side the staging
geometry implies (often its back). Show the functional side to the
camera ONLY when the shot text itself stages it toward the viewer.
Never flip, mirror or reverse an object's front and back.

CARRIED STATE (persist exactly — must match the neighbouring shots of this scene): Min-suk remains dead. She is seated motionless on the floor behind the curtain with her torso slumped and her head dropped forward; one rigid hand is closed around a crumpled photograph. The circular wrist mark is present and turns red as Su-ri-young removes the photograph.

NO PEOPLE IN THIS SHOT: the shot text shows only the place and its state — no living person or any body part appears in frame, unless the shot text itself explicitly says so.

No text, captions, watermarks or annotations anywhere.

Create ONE FINAL photorealistic live-action film still of the moment below — contemporary South Korea, 2026; all people are Korean unless stated. TIME OF DAY (lock): night.

SHOT TEXT (authoritative, Korean): 죽은 강민숙의 굳어버린 손안에 구겨진 종이 사진이 쥐어져 있는 클로즈업

LOCATION (lock): On the floor behind the curtain in the daughter's bedroom, beside the dead mother's body in the dim nighttime apartment. The shot takes place here — the attached PREVIOUS SHOT STILL shows this exact place.

CAMERA & FRAME (follow exactly — this is the composition authority for this still; the reference images supply identity and place, never the framing):
- CAMERA: At hand height, a steep but not overhead downward angle isolates 강민숙’s rigid fingers and the crumpled photograph while retaining a narrow strip of her forearm and the nearby floor for scale. The hand lies across the lower-center and the photograph remains below forty percent of the frame, its creased image side caught between the closed fingers.
- FRAMING SCALE: insert close-up on a detail
- KEY BACKGROUND ELEMENTS: 구겨진 종이 사진 (Crumpled and gripped inside the hand) — The camera sees the crumpled image-bearing face, partially concealed by fingers and folds; used as Central clue held within the rigid hand; 바닥 (No additional state specified in this framing); used as Scale reference at the edge of the isolated detail.
Compose the frame exactly as specified above — camera angle, subject scale and screen placement. Keep true physical scale between people and background elements: fixtures and distant objects occupy only the small screen area their real size and distance dictate; never enlarge a background object into a foreground presence.

THIS SHOT CONTINUES THE PREVIOUS SHOT: everything the attached still established — the place, its fixed features and wear, each person's clothing and state — persists. Any person in it who cannot move stays PRECISELY as photographed (body, pose, contact points, held objects); only the camera changes.

REALIZE FIGURATIVE LANGUAGE AS A LIVE-ACTION SHOT: the Korean shot
text may describe characters metaphorically, figuratively or with
exaggeration. Photograph what a real movie camera would actually
record on a physical set — exaggerated or figurative impressions
become realistic staging choices (lighting, distance, angle,
wardrobe), not literal fantasy imagery.
EVERY CHARACTER IS A HUMAN BEING: unless the story explicitly
features non-human or virtual beings (as in science fiction or
fantasy), every character — however indirectly, vaguely or
figuratively the text describes them — IS a real human. When the
text gives no direct visual description of a person, IMAGINE one
and still show them as a concrete, fully-formed human being:
always render the human form to the maximum extent — build,
posture, face, hands, clothing — never reduce a person to a shape,
blob, solid silhouette or abstract mass.

EXPRESSIONS ARE ACTED, NEVER ANATOMICAL: when the text describes a
person's eyes, face or presence emotionally or figuratively (vacant,
hollow, dazed, burning, lifeless gaze and the like), it describes an
ACTOR'S PERFORMANCE captured by a real camera — realize it ONLY
through gaze direction, focus, eyelids, facial muscles, stillness
and posture. Every living person's eyes remain anatomically normal
human eyes with a natural iris and pupil, natural sclera and normal
proportions; NEVER whiten, blank out, cloud over, glow, enlarge or
otherwise alter eyeballs, skin or anatomy — unless the story
explicitly declares that being non-human or supernatural in form.

PROPS FACE THE RIGHT WAY: every handheld or used object must be
oriented exactly as its real-world use requires. A person reading,
watching or operating something (a phone, a photograph, a paper,
any device) has its functional side — screen, front, page — facing
THEIR OWN eyes; the camera then sees whatever side the staging
geometry implies (often its back). Show the functional side to the
camera ONLY when the shot text itself stages it toward the viewer.
Never flip, mirror or reverse an object's front and back.

CARRIED STATE (persist exactly — must match the neighbouring shots of this scene): Min-suk remains dead. She is seated motionless on the floor behind the curtain with her torso slumped and her head dropped forward; one rigid hand is closed around a crumpled photograph. The circular wrist mark is present and turns red as Su-ri-young removes the photograph.

NO PEOPLE IN THIS SHOT: the shot text shows only the place and its state — no living person or any body part appears in frame, unless the shot text itself explicitly says so.

No text, captions, watermarks or annotations anywhere.

VISUAL APPROACH — this take (camera, framing, composition):
Use a restrained, forensic visual treatment with crisp focus carried across the rigid fingers, crushed paper folds, and the narrow visible strip of forearm. Let soft, cool nighttime illumination fall obliquely across the hand, creating controlled micro-shadows in the knuckles and creases while the nearby floor recedes quietly; keep the image-bearing face of the photograph legible only to the extent allowed by its concealment and folds.

PRIORITY OF INSTRUCTIONS: the VISUAL APPROACH above governs camera,
framing and composition ONLY. Every other instruction in this prompt —
location, time of day, people and their identity, poses, props,
carried state, physical rules, and the no-text rule — is binding
exactly as written earlier. If the VISUAL APPROACH appears to conflict
with any of them, the earlier instruction wins.

Create ONE FINAL photorealistic live-action film still of the moment below — contemporary South Korea, 2026; all people are Korean unless stated. TIME OF DAY (lock): night.

SHOT TEXT (authoritative, Korean): 죽은 강민숙의 굳어버린 손안에 구겨진 종이 사진이 쥐어져 있는 클로즈업

LOCATION (lock): On the floor behind the curtain in the daughter's bedroom, beside the dead mother's body in the dim nighttime apartment. The shot takes place here — the attached PREVIOUS SHOT STILL shows this exact place.

CAMERA & FRAME (follow exactly — this is the composition authority for this still; the reference images supply identity and place, never the framing):
- CAMERA: At hand height, a steep but not overhead downward angle isolates 강민숙’s rigid fingers and the crumpled photograph while retaining a narrow strip of her forearm and the nearby floor for scale. The hand lies across the lower-center and the photograph remains below forty percent of the frame, its creased image side caught between the closed fingers.
- FRAMING SCALE: insert close-up on a detail
- KEY BACKGROUND ELEMENTS: 구겨진 종이 사진 (Crumpled and gripped inside the hand) — The camera sees the crumpled image-bearing face, partially concealed by fingers and folds; used as Central clue held within the rigid hand; 바닥 (No additional state specified in this framing); used as Scale reference at the edge of the isolated detail.
Compose the frame exactly as specified above — camera angle, subject scale and screen placement. Keep true physical scale between people and background elements: fixtures and distant objects occupy only the small screen area their real size and distance dictate; never enlarge a background object into a foreground presence.

THIS SHOT CONTINUES THE PREVIOUS SHOT: everything the attached still established — the place, its fixed features and wear, each person's clothing and state — persists. Any person in it who cannot move stays PRECISELY as photographed (body, pose, contact points, held objects); only the camera changes.

REALIZE FIGURATIVE LANGUAGE AS A LIVE-ACTION SHOT: the Korean shot
text may describe characters metaphorically, figuratively or with
exaggeration. Photograph what a real movie camera would actually
record on a physical set — exaggerated or figurative impressions
become realistic staging choices (lighting, distance, angle,
wardrobe), not literal fantasy imagery.
EVERY CHARACTER IS A HUMAN BEING: unless the story explicitly
features non-human or virtual beings (as in science fiction or
fantasy), every character — however indirectly, vaguely or
figuratively the text describes them — IS a real human. When the
text gives no direct visual description of a person, IMAGINE one
and still show them as a concrete, fully-formed human being:
always render the human form to the maximum extent — build,
posture, face, hands, clothing — never reduce a person to a shape,
blob, solid silhouette or abstract mass.

EXPRESSIONS ARE ACTED, NEVER ANATOMICAL: when the text describes a
person's eyes, face or presence emotionally or figuratively (vacant,
hollow, dazed, burning, lifeless gaze and the like), it describes an
ACTOR'S PERFORMANCE captured by a real camera — realize it ONLY
through gaze direction, focus, eyelids, facial muscles, stillness
and posture. Every living person's eyes remain anatomically normal
human eyes with a natural iris and pupil, natural sclera and normal
proportions; NEVER whiten, blank out, cloud over, glow, enlarge or
otherwise alter eyeballs, skin or anatomy — unless the story
explicitly declares that being non-human or supernatural in form.

PROPS FACE THE RIGHT WAY: every handheld or used object must be
oriented exactly as its real-world use requires. A person reading,
watching or operating something (a phone, a photograph, a paper,
any device) has its functional side — screen, front, page — facing
THEIR OWN eyes; the camera then sees whatever side the staging
geometry implies (often its back). Show the functional side to the
camera ONLY when the shot text itself stages it toward the viewer.
Never flip, mirror or reverse an object's front and back.

CARRIED STATE (persist exactly — must match the neighbouring shots of this scene): Min-suk remains dead. She is seated motionless on the floor behind the curtain with her torso slumped and her head dropped forward; one rigid hand is closed around a crumpled photograph. The circular wrist mark is present and turns red as Su-ri-young removes the photograph.

NO PEOPLE IN THIS SHOT: the shot text shows only the place and its state — no living person or any body part appears in frame, unless the shot text itself explicitly says so.

No text, captions, watermarks or annotations anywhere.

VISUAL APPROACH — this take (camera, framing, composition):
Render the insert with very shallow, tactile focus concentrated on the pinched edge of the crumpled photograph and the closed fingertips, allowing the nearer forearm edge and surrounding floor to dissolve gently. Use dim raking nighttime light to catch isolated highlights along the paper ridges and rigid finger contours, with restrained shadow falloff making the photograph feel trapped inside the hand.

PRIORITY OF INSTRUCTIONS: the VISUAL APPROACH above governs camera,
framing and composition ONLY. Every other instruction in this prompt —
location, time of day, people and their identity, poses, props,
carried state, physical rules, and the no-text rule — is binding
exactly as written earlier. If the VISUAL APPROACH appears to conflict
with any of them, the earlier instruction wins.

Edit this photograph. Apply ONLY the corrections listed below,
each in place, changing nothing else:

CORRECTIONS:
- Lift the forearm off the floor so it angles downwards with only the hand making contact with the ground.
- Change the camera angle to a steep downward view pointing at the hand and floor.
- Crumple the photograph into a much tighter ball gripped mostly inside the closed fist, matching the reference.

PRESERVE EVERYTHING ELSE EXACTLY as in the original photograph —
every person, pose, face, garment, object, set feature, lighting,
camera and framing stay identical. No text, captions or
watermarks anywhere.

Create ONE FINAL photorealistic live-action film still of the moment below — contemporary South Korea, 2026; all people are Korean unless stated. TIME OF DAY (lock): night.

SHOT TEXT (authoritative, Korean): 침대 위쪽 벽면에 붉은색의 굵은 원 형태가 선명하게 남아 있는 모습

LOCATION (lock): Inside the daughter's bedroom, on the wall directly above the bed in the dim nighttime apartment. The shot takes place here — the attached PREVIOUS SHOT STILL shows this exact place.

CAMERA & FRAME (follow exactly — this is the composition authority for this still; the reference images supply identity and place, never the framing):
- CAMERA: From low bedside height, the upward tilt passes the lower room and settles into a static, slightly low-angle view of the wall above the bed. The complete thick red circle sits in the upper-center at a restrained scale, with the bed edge retained along the lower frame to establish its physical placement rather than turning the mark into an abstract graphic.
- FRAMING SCALE: medium shot
- FRAME LAYOUT: wall with thick red circle above the bed in the upper-center of the frame, background; bed beneath the circle in the lower-center of the frame, midground.
- KEY BACKGROUND ELEMENTS: 침대 위 벽면 (Marked by a clearly visible red circle) — The camera-facing wall plane bears a complete thick red circle above the bed; used as Final supernatural focal surface; 침대 (Present beneath the circle) — Its upper edge is seen below the marked wall from the low bedside angle; used as Lower-frame spatial reference locating the mark.
Compose the frame exactly as specified above — camera angle, subject scale and screen placement. Keep true physical scale between people and background elements: fixtures and distant objects occupy only the small screen area their real size and distance dictate; never enlarge a background object into a foreground presence.

THIS SHOT CONTINUES THE PREVIOUS SHOT: everything the attached still established — the place, its fixed features and wear, each person's clothing and state — persists. Any person in it who cannot move stays PRECISELY as photographed (body, pose, contact points, held objects); only the camera changes.

REALIZE FIGURATIVE LANGUAGE AS A LIVE-ACTION SHOT: the Korean shot
text may describe characters metaphorically, figuratively or with
exaggeration. Photograph what a real movie camera would actually
record on a physical set — exaggerated or figurative impressions
become realistic staging choices (lighting, distance, angle,
wardrobe), not literal fantasy imagery.
EVERY CHARACTER IS A HUMAN BEING: unless the story explicitly
features non-human or virtual beings (as in science fiction or
fantasy), every character — however indirectly, vaguely or
figuratively the text describes them — IS a real human. When the
text gives no direct visual description of a person, IMAGINE one
and still show them as a concrete, fully-formed human being:
always render the human form to the maximum extent — build,
posture, face, hands, clothing — never reduce a person to a shape,
blob, solid silhouette or abstract mass.

EXPRESSIONS ARE ACTED, NEVER ANATOMICAL: when the text describes a
person's eyes, face or presence emotionally or figuratively (vacant,
hollow, dazed, burning, lifeless gaze and the like), it describes an
ACTOR'S PERFORMANCE captured by a real camera — realize it ONLY
through gaze direction, focus, eyelids, facial muscles, stillness
and posture. Every living person's eyes remain anatomically normal
human eyes with a natural iris and pupil, natural sclera and normal
proportions; NEVER whiten, blank out, cloud over, glow, enlarge or
otherwise alter eyeballs, skin or anatomy — unless the story
explicitly declares that being non-human or supernatural in form.

PROPS FACE THE RIGHT WAY: every handheld or used object must be
oriented exactly as its real-world use requires. A person reading,
watching or operating something (a phone, a photograph, a paper,
any device) has its functional side — screen, front, page — facing
THEIR OWN eyes; the camera then sees whatever side the staging
geometry implies (often its back). Show the functional side to the
camera ONLY when the shot text itself stages it toward the viewer.
Never flip, mirror or reverse an object's front and back.

CARRIED STATE (persist exactly — must match the neighbouring shots of this scene): The thick red circle now remains visible on the wall above the bed; Min-suk’s seated body, wounds, blood, and bloody footprints are still present in the room. Su-ri-young has taken the crumpled photograph from Min-suk’s hand.

NO PEOPLE IN THIS SHOT: the shot text shows only the place and its state — no living person or any body part appears in frame, unless the shot text itself explicitly says so.

No text, captions, watermarks or annotations anywhere.

Create ONE FINAL photorealistic live-action film still of the moment below — contemporary South Korea, 2026; all people are Korean unless stated. TIME OF DAY (lock): night.

SHOT TEXT (authoritative, Korean): 침대 위쪽 벽면에 붉은색의 굵은 원 형태가 선명하게 남아 있는 모습

LOCATION (lock): Inside the daughter's bedroom, on the wall directly above the bed in the dim nighttime apartment. The shot takes place here — the attached PREVIOUS SHOT STILL shows this exact place.

CAMERA & FRAME (follow exactly — this is the composition authority for this still; the reference images supply identity and place, never the framing):
- CAMERA: From low bedside height, the upward tilt passes the lower room and settles into a static, slightly low-angle view of the wall above the bed. The complete thick red circle sits in the upper-center at a restrained scale, with the bed edge retained along the lower frame to establish its physical placement rather than turning the mark into an abstract graphic.
- FRAMING SCALE: medium shot
- FRAME LAYOUT: wall with thick red circle above the bed in the upper-center of the frame, background; bed beneath the circle in the lower-center of the frame, midground.
- KEY BACKGROUND ELEMENTS: 침대 위 벽면 (Marked by a clearly visible red circle) — The camera-facing wall plane bears a complete thick red circle above the bed; used as Final supernatural focal surface; 침대 (Present beneath the circle) — Its upper edge is seen below the marked wall from the low bedside angle; used as Lower-frame spatial reference locating the mark.
Compose the frame exactly as specified above — camera angle, subject scale and screen placement. Keep true physical scale between people and background elements: fixtures and distant objects occupy only the small screen area their real size and distance dictate; never enlarge a background object into a foreground presence.

THIS SHOT CONTINUES THE PREVIOUS SHOT: everything the attached still established — the place, its fixed features and wear, each person's clothing and state — persists. Any person in it who cannot move stays PRECISELY as photographed (body, pose, contact points, held objects); only the camera changes.

REALIZE FIGURATIVE LANGUAGE AS A LIVE-ACTION SHOT: the Korean shot
text may describe characters metaphorically, figuratively or with
exaggeration. Photograph what a real movie camera would actually
record on a physical set — exaggerated or figurative impressions
become realistic staging choices (lighting, distance, angle,
wardrobe), not literal fantasy imagery.
EVERY CHARACTER IS A HUMAN BEING: unless the story explicitly
features non-human or virtual beings (as in science fiction or
fantasy), every character — however indirectly, vaguely or
figuratively the text describes them — IS a real human. When the
text gives no direct visual description of a person, IMAGINE one
and still show them as a concrete, fully-formed human being:
always render the human form to the maximum extent — build,
posture, face, hands, clothing — never reduce a person to a shape,
blob, solid silhouette or abstract mass.

EXPRESSIONS ARE ACTED, NEVER ANATOMICAL: when the text describes a
person's eyes, face or presence emotionally or figuratively (vacant,
hollow, dazed, burning, lifeless gaze and the like), it describes an
ACTOR'S PERFORMANCE captured by a real camera — realize it ONLY
through gaze direction, focus, eyelids, facial muscles, stillness
and posture. Every living person's eyes remain anatomically normal
human eyes with a natural iris and pupil, natural sclera and normal
proportions; NEVER whiten, blank out, cloud over, glow, enlarge or
otherwise alter eyeballs, skin or anatomy — unless the story
explicitly declares that being non-human or supernatural in form.

PROPS FACE THE RIGHT WAY: every handheld or used object must be
oriented exactly as its real-world use requires. A person reading,
watching or operating something (a phone, a photograph, a paper,
any device) has its functional side — screen, front, page — facing
THEIR OWN eyes; the camera then sees whatever side the staging
geometry implies (often its back). Show the functional side to the
camera ONLY when the shot text itself stages it toward the viewer.
Never flip, mirror or reverse an object's front and back.

CARRIED STATE (persist exactly — must match the neighbouring shots of this scene): The thick red circle now remains visible on the wall above the bed; Min-suk’s seated body, wounds, blood, and bloody footprints are still present in the room. Su-ri-young has taken the crumpled photograph from Min-suk’s hand.

NO PEOPLE IN THIS SHOT: the shot text shows only the place and its state — no living person or any body part appears in frame, unless the shot text itself explicitly says so.

No text, captions, watermarks or annotations anywhere.

VISUAL APPROACH — this take (camera, framing, composition):
Use an austere, nearly frontal composition with strong vertical and horizontal geometry, preserving the complete red circle at restrained scale in the upper center and the bed edge as a firm lower-frame anchor. Hold both wall texture and bed edge in clear focus under dim, soft nighttime illumination, with subdued surrounding color and controlled contrast so the red mark remains vivid without appearing luminous or graphic.

PRIORITY OF INSTRUCTIONS: the VISUAL APPROACH above governs camera,
framing and composition ONLY. Every other instruction in this prompt —
location, time of day, people and their identity, poses, props,
carried state, physical rules, and the no-text rule — is binding
exactly as written earlier. If the VISUAL APPROACH appears to conflict
with any of them, the earlier instruction wins.

Create ONE FINAL photorealistic live-action film still of the moment below — contemporary South Korea, 2026; all people are Korean unless stated. TIME OF DAY (lock): night.

SHOT TEXT (authoritative, Korean): 침대 위쪽 벽면에 붉은색의 굵은 원 형태가 선명하게 남아 있는 모습

LOCATION (lock): Inside the daughter's bedroom, on the wall directly above the bed in the dim nighttime apartment. The shot takes place here — the attached PREVIOUS SHOT STILL shows this exact place.

CAMERA & FRAME (follow exactly — this is the composition authority for this still; the reference images supply identity and place, never the framing):
- CAMERA: From low bedside height, the upward tilt passes the lower room and settles into a static, slightly low-angle view of the wall above the bed. The complete thick red circle sits in the upper-center at a restrained scale, with the bed edge retained along the lower frame to establish its physical placement rather than turning the mark into an abstract graphic.
- FRAMING SCALE: medium shot
- FRAME LAYOUT: wall with thick red circle above the bed in the upper-center of the frame, background; bed beneath the circle in the lower-center of the frame, midground.
- KEY BACKGROUND ELEMENTS: 침대 위 벽면 (Marked by a clearly visible red circle) — The camera-facing wall plane bears a complete thick red circle above the bed; used as Final supernatural focal surface; 침대 (Present beneath the circle) — Its upper edge is seen below the marked wall from the low bedside angle; used as Lower-frame spatial reference locating the mark.
Compose the frame exactly as specified above — camera angle, subject scale and screen placement. Keep true physical scale between people and background elements: fixtures and distant objects occupy only the small screen area their real size and distance dictate; never enlarge a background object into a foreground presence.

THIS SHOT CONTINUES THE PREVIOUS SHOT: everything the attached still established — the place, its fixed features and wear, each person's clothing and state — persists. Any person in it who cannot move stays PRECISELY as photographed (body, pose, contact points, held objects); only the camera changes.

REALIZE FIGURATIVE LANGUAGE AS A LIVE-ACTION SHOT: the Korean shot
text may describe characters metaphorically, figuratively or with
exaggeration. Photograph what a real movie camera would actually
record on a physical set — exaggerated or figurative impressions
become realistic staging choices (lighting, distance, angle,
wardrobe), not literal fantasy imagery.
EVERY CHARACTER IS A HUMAN BEING: unless the story explicitly
features non-human or virtual beings (as in science fiction or
fantasy), every character — however indirectly, vaguely or
figuratively the text describes them — IS a real human. When the
text gives no direct visual description of a person, IMAGINE one
and still show them as a concrete, fully-formed human being:
always render the human form to the maximum extent — build,
posture, face, hands, clothing — never reduce a person to a shape,
blob, solid silhouette or abstract mass.

EXPRESSIONS ARE ACTED, NEVER ANATOMICAL: when the text describes a
person's eyes, face or presence emotionally or figuratively (vacant,
hollow, dazed, burning, lifeless gaze and the like), it describes an
ACTOR'S PERFORMANCE captured by a real camera — realize it ONLY
through gaze direction, focus, eyelids, facial muscles, stillness
and posture. Every living person's eyes remain anatomically normal
human eyes with a natural iris and pupil, natural sclera and normal
proportions; NEVER whiten, blank out, cloud over, glow, enlarge or
otherwise alter eyeballs, skin or anatomy — unless the story
explicitly declares that being non-human or supernatural in form.

PROPS FACE THE RIGHT WAY: every handheld or used object must be
oriented exactly as its real-world use requires. A person reading,
watching or operating something (a phone, a photograph, a paper,
any device) has its functional side — screen, front, page — facing
THEIR OWN eyes; the camera then sees whatever side the staging
geometry implies (often its back). Show the functional side to the
camera ONLY when the shot text itself stages it toward the viewer.
Never flip, mirror or reverse an object's front and back.

CARRIED STATE (persist exactly — must match the neighbouring shots of this scene): The thick red circle now remains visible on the wall above the bed; Min-suk’s seated body, wounds, blood, and bloody footprints are still present in the room. Su-ri-young has taken the crumpled photograph from Min-suk’s hand.

NO PEOPLE IN THIS SHOT: the shot text shows only the place and its state — no living person or any body part appears in frame, unless the shot text itself explicitly says so.

No text, captions, watermarks or annotations anywhere.

VISUAL APPROACH — this take (camera, framing, composition):
Photograph from the locked low bedside angle with selective focus: render the wall and complete red circle sharply while allowing the lower bed edge to fall gently soft, still clearly readable as the spatial reference. Shape the dim nighttime light so it grazes the wall plane, revealing subtle surface texture and the physical thickness of the red mark; let the lower room recede into soft shadow for a tense, intimate sense of discovery.

PRIORITY OF INSTRUCTIONS: the VISUAL APPROACH above governs camera,
framing and composition ONLY. Every other instruction in this prompt —
location, time of day, people and their identity, poses, props,
carried state, physical rules, and the no-text rule — is binding
exactly as written earlier. If the VISUAL APPROACH appears to conflict
with any of them, the earlier instruction wins.

Create ONE FINAL photorealistic live-action film still of the moment below — contemporary South Korea, 2026; all people are Korean unless stated. TIME OF DAY (lock): night.

SHOT TEXT (authoritative, Korean): 핏자국 하나 없이 깨끗한 바닥과 빈 커튼 쪽이 보이는 수리영 방안 전경

LOCATION (lock): Inside the daughter's unnaturally tidy bedroom, facing the spotless floor and the empty space behind the curtain. The shot takes place here — the attached LOCATION PHOTOGRAPH shows the exact spot.

CAMERA & FRAME (follow exactly — this is the composition authority for this still; the reference images supply identity and place, never the framing):
- CAMERA: From eye level at the room doorway, hold a wide, slightly downward view along the flow axis, placing the clean floor across the lower half, the bed to one side, and the empty curtain area in the rear of frame. The static composition withholds any human presence and uses the doorway edges as a restrained frame, preparing for a straight push into the room.
- FRAMING SCALE: wide shot
- FRAME LAYOUT: clean floor without blood in the lower-center of the frame, foreground; empty curtain area in the upper-right of the frame, background.
- KEY BACKGROUND ELEMENTS: bedroom floor (Clean and showing no bloodstains); used as Occupies the lower half of the composition as the primary evidence of absence; bed (Neatly arranged with no red mark visible) — Its side and upper surface are visible from the doorway; used as Provides spatial context beside the cleared floor; curtain area (No body remains behind the curtain) — The room-facing side is visible, with the area behind it presented as empty; used as Forms the rear visual endpoint where the body had been seen; bedroom doorway (Open) — The inner sides face the camera from its threshold position; used as Creates a narrow foreground frame around the room-wide reveal.
Compose the frame exactly as specified above — camera angle, subject scale and screen placement. Keep true physical scale between people and background elements: fixtures and distant objects occupy only the small screen area their real size and distance dictate; never enlarge a background object into a foreground presence.

REALIZE FIGURATIVE LANGUAGE AS A LIVE-ACTION SHOT: the Korean shot
text may describe characters metaphorically, figuratively or with
exaggeration. Photograph what a real movie camera would actually
record on a physical set — exaggerated or figurative impressions
become realistic staging choices (lighting, distance, angle,
wardrobe), not literal fantasy imagery.
EVERY CHARACTER IS A HUMAN BEING: unless the story explicitly
features non-human or virtual beings (as in science fiction or
fantasy), every character — however indirectly, vaguely or
figuratively the text describes them — IS a real human. When the
text gives no direct visual description of a person, IMAGINE one
and still show them as a concrete, fully-formed human being:
always render the human form to the maximum extent — build,
posture, face, hands, clothing — never reduce a person to a shape,
blob, solid silhouette or abstract mass.

EXPRESSIONS ARE ACTED, NEVER ANATOMICAL: when the text describes a
person's eyes, face or presence emotionally or figuratively (vacant,
hollow, dazed, burning, lifeless gaze and the like), it describes an
ACTOR'S PERFORMANCE captured by a real camera — realize it ONLY
through gaze direction, focus, eyelids, facial muscles, stillness
and posture. Every living person's eyes remain anatomically normal
human eyes with a natural iris and pupil, natural sclera and normal
proportions; NEVER whiten, blank out, cloud over, glow, enlarge or
otherwise alter eyeballs, skin or anatomy — unless the story
explicitly declares that being non-human or supernatural in form.

PROPS FACE THE RIGHT WAY: every handheld or used object must be
oriented exactly as its real-world use requires. A person reading,
watching or operating something (a phone, a photograph, a paper,
any device) has its functional side — screen, front, page — facing
THEIR OWN eyes; the camera then sees whatever side the staging
geometry implies (often its back). Show the functional side to the
camera ONLY when the shot text itself stages it toward the viewer.
Never flip, mirror or reverse an object's front and back.

CARRIED STATE (persist exactly — must match the neighbouring shots of this scene): The room is now unnaturally clean: Min-suk’s body, all blood and bloody footprints, and the red wall circle have disappeared, leaving the curtain area empty.

NO PEOPLE IN THIS SHOT: the shot text shows only the place and its state — no living person or any body part appears in frame, unless the shot text itself explicitly says so.

No text, captions, watermarks or annotations anywhere.

Create ONE FINAL photorealistic live-action film still of the moment below — contemporary South Korea, 2026; all people are Korean unless stated. TIME OF DAY (lock): night.

SHOT TEXT (authoritative, Korean): 핏자국 하나 없이 깨끗한 바닥과 빈 커튼 쪽이 보이는 수리영 방안 전경

LOCATION (lock): Inside the daughter's unnaturally tidy bedroom, facing the spotless floor and the empty space behind the curtain. The shot takes place here — the attached LOCATION PHOTOGRAPH shows the exact spot.

CAMERA & FRAME (follow exactly — this is the composition authority for this still; the reference images supply identity and place, never the framing):
- CAMERA: From eye level at the room doorway, hold a wide, slightly downward view along the flow axis, placing the clean floor across the lower half, the bed to one side, and the empty curtain area in the rear of frame. The static composition withholds any human presence and uses the doorway edges as a restrained frame, preparing for a straight push into the room.
- FRAMING SCALE: wide shot
- FRAME LAYOUT: clean floor without blood in the lower-center of the frame, foreground; empty curtain area in the upper-right of the frame, background.
- KEY BACKGROUND ELEMENTS: bedroom floor (Clean and showing no bloodstains); used as Occupies the lower half of the composition as the primary evidence of absence; bed (Neatly arranged with no red mark visible) — Its side and upper surface are visible from the doorway; used as Provides spatial context beside the cleared floor; curtain area (No body remains behind the curtain) — The room-facing side is visible, with the area behind it presented as empty; used as Forms the rear visual endpoint where the body had been seen; bedroom doorway (Open) — The inner sides face the camera from its threshold position; used as Creates a narrow foreground frame around the room-wide reveal.
Compose the frame exactly as specified above — camera angle, subject scale and screen placement. Keep true physical scale between people and background elements: fixtures and distant objects occupy only the small screen area their real size and distance dictate; never enlarge a background object into a foreground presence.

REALIZE FIGURATIVE LANGUAGE AS A LIVE-ACTION SHOT: the Korean shot
text may describe characters metaphorically, figuratively or with
exaggeration. Photograph what a real movie camera would actually
record on a physical set — exaggerated or figurative impressions
become realistic staging choices (lighting, distance, angle,
wardrobe), not literal fantasy imagery.
EVERY CHARACTER IS A HUMAN BEING: unless the story explicitly
features non-human or virtual beings (as in science fiction or
fantasy), every character — however indirectly, vaguely or
figuratively the text describes them — IS a real human. When the
text gives no direct visual description of a person, IMAGINE one
and still show them as a concrete, fully-formed human being:
always render the human form to the maximum extent — build,
posture, face, hands, clothing — never reduce a person to a shape,
blob, solid silhouette or abstract mass.

EXPRESSIONS ARE ACTED, NEVER ANATOMICAL: when the text describes a
person's eyes, face or presence emotionally or figuratively (vacant,
hollow, dazed, burning, lifeless gaze and the like), it describes an
ACTOR'S PERFORMANCE captured by a real camera — realize it ONLY
through gaze direction, focus, eyelids, facial muscles, stillness
and posture. Every living person's eyes remain anatomically normal
human eyes with a natural iris and pupil, natural sclera and normal
proportions; NEVER whiten, blank out, cloud over, glow, enlarge or
otherwise alter eyeballs, skin or anatomy — unless the story
explicitly declares that being non-human or supernatural in form.

PROPS FACE THE RIGHT WAY: every handheld or used object must be
oriented exactly as its real-world use requires. A person reading,
watching or operating something (a phone, a photograph, a paper,
any device) has its functional side — screen, front, page — facing
THEIR OWN eyes; the camera then sees whatever side the staging
geometry implies (often its back). Show the functional side to the
camera ONLY when the shot text itself stages it toward the viewer.
Never flip, mirror or reverse an object's front and back.

CARRIED STATE (persist exactly — must match the neighbouring shots of this scene): The room is now unnaturally clean: Min-suk’s body, all blood and bloody footprints, and the red wall circle have disappeared, leaving the curtain area empty.

NO PEOPLE IN THIS SHOT: the shot text shows only the place and its state — no living person or any body part appears in frame, unless the shot text itself explicitly says so.

No text, captions, watermarks or annotations anywhere.

VISUAL APPROACH — this take (camera, framing, composition):
Render the doorway-wide view with restrained, cool nighttime illumination and deep focus, keeping the spotless foreground floor, neatly arranged bed, and empty curtain area equally legible. Favor rigid lines, balanced negative space, and low contrast so the room’s unnatural cleanliness reads with clinical clarity; the doorway edges remain quiet, dark framing elements rather than a dominant silhouette.

PRIORITY OF INSTRUCTIONS: the VISUAL APPROACH above governs camera,
framing and composition ONLY. Every other instruction in this prompt —
location, time of day, people and their identity, poses, props,
carried state, physical rules, and the no-text rule — is binding
exactly as written earlier. If the VISUAL APPROACH appears to conflict
with any of them, the earlier instruction wins.

Create ONE FINAL photorealistic live-action film still of the moment below — contemporary South Korea, 2026; all people are Korean unless stated. TIME OF DAY (lock): night.

SHOT TEXT (authoritative, Korean): 핏자국 하나 없이 깨끗한 바닥과 빈 커튼 쪽이 보이는 수리영 방안 전경

LOCATION (lock): Inside the daughter's unnaturally tidy bedroom, facing the spotless floor and the empty space behind the curtain. The shot takes place here — the attached LOCATION PHOTOGRAPH shows the exact spot.

CAMERA & FRAME (follow exactly — this is the composition authority for this still; the reference images supply identity and place, never the framing):
- CAMERA: From eye level at the room doorway, hold a wide, slightly downward view along the flow axis, placing the clean floor across the lower half, the bed to one side, and the empty curtain area in the rear of frame. The static composition withholds any human presence and uses the doorway edges as a restrained frame, preparing for a straight push into the room.
- FRAMING SCALE: wide shot
- FRAME LAYOUT: clean floor without blood in the lower-center of the frame, foreground; empty curtain area in the upper-right of the frame, background.
- KEY BACKGROUND ELEMENTS: bedroom floor (Clean and showing no bloodstains); used as Occupies the lower half of the composition as the primary evidence of absence; bed (Neatly arranged with no red mark visible) — Its side and upper surface are visible from the doorway; used as Provides spatial context beside the cleared floor; curtain area (No body remains behind the curtain) — The room-facing side is visible, with the area behind it presented as empty; used as Forms the rear visual endpoint where the body had been seen; bedroom doorway (Open) — The inner sides face the camera from its threshold position; used as Creates a narrow foreground frame around the room-wide reveal.
Compose the frame exactly as specified above — camera angle, subject scale and screen placement. Keep true physical scale between people and background elements: fixtures and distant objects occupy only the small screen area their real size and distance dictate; never enlarge a background object into a foreground presence.

REALIZE FIGURATIVE LANGUAGE AS A LIVE-ACTION SHOT: the Korean shot
text may describe characters metaphorically, figuratively or with
exaggeration. Photograph what a real movie camera would actually
record on a physical set — exaggerated or figurative impressions
become realistic staging choices (lighting, distance, angle,
wardrobe), not literal fantasy imagery.
EVERY CHARACTER IS A HUMAN BEING: unless the story explicitly
features non-human or virtual beings (as in science fiction or
fantasy), every character — however indirectly, vaguely or
figuratively the text describes them — IS a real human. When the
text gives no direct visual description of a person, IMAGINE one
and still show them as a concrete, fully-formed human being:
always render the human form to the maximum extent — build,
posture, face, hands, clothing — never reduce a person to a shape,
blob, solid silhouette or abstract mass.

EXPRESSIONS ARE ACTED, NEVER ANATOMICAL: when the text describes a
person's eyes, face or presence emotionally or figuratively (vacant,
hollow, dazed, burning, lifeless gaze and the like), it describes an
ACTOR'S PERFORMANCE captured by a real camera — realize it ONLY
through gaze direction, focus, eyelids, facial muscles, stillness
and posture. Every living person's eyes remain anatomically normal
human eyes with a natural iris and pupil, natural sclera and normal
proportions; NEVER whiten, blank out, cloud over, glow, enlarge or
otherwise alter eyeballs, skin or anatomy — unless the story
explicitly declares that being non-human or supernatural in form.

PROPS FACE THE RIGHT WAY: every handheld or used object must be
oriented exactly as its real-world use requires. A person reading,
watching or operating something (a phone, a photograph, a paper,
any device) has its functional side — screen, front, page — facing
THEIR OWN eyes; the camera then sees whatever side the staging
geometry implies (often its back). Show the functional side to the
camera ONLY when the shot text itself stages it toward the viewer.
Never flip, mirror or reverse an object's front and back.

CARRIED STATE (persist exactly — must match the neighbouring shots of this scene): The room is now unnaturally clean: Min-suk’s body, all blood and bloody footprints, and the red wall circle have disappeared, leaving the curtain area empty.

NO PEOPLE IN THIS SHOT: the shot text shows only the place and its state — no living person or any body part appears in frame, unless the shot text itself explicitly says so.

No text, captions, watermarks or annotations anywhere.

VISUAL APPROACH — this take (camera, framing, composition):
Let the doorway edges fall into heavier shadow while soft nighttime light recedes across the clean floor toward the empty curtain area, creating a tense visual corridor into the room. Use layered depth and gentle contrast, with the bed held as a subdued side mass and the vacant rear space as the composition’s unsettling endpoint; preserve enough focus throughout to verify that no blood, body, or red mark remains.

PRIORITY OF INSTRUCTIONS: the VISUAL APPROACH above governs camera,
framing and composition ONLY. Every other instruction in this prompt —
location, time of day, people and their identity, poses, props,
carried state, physical rules, and the no-text rule — is binding
exactly as written earlier. If the VISUAL APPROACH appears to conflict
with any of them, the earlier instruction wins.

Create ONE FINAL photorealistic live-action film still of the moment below — contemporary South Korea, 2026; all people are Korean unless stated. TIME OF DAY (lock): dawn.

SHOT TEXT (authoritative, Korean): 새벽의 푸르스름한 빛이 감도는 옥탑방 외벽을 배경으로 검은 고양이가 고개를 돌려 창문을 바라보는 모습

LOCATION (lock): Outside on the villa rooftop at dawn, beside the rooftop apartment's wall and its illuminated window. The shot takes place here — the attached LOCATION PHOTOGRAPH shows the exact spot.

STRUCTURE LOOK AUTHORITY: the attached STRUCTURE LOOK photograph is the identity of the fixed structure at this location — wherever that structure appears in the frame, its shape, proportions, openings, materials and colors are LOCKED to it. The LOCATION PHOTOGRAPH remains the authority for this shot's sub-space, surroundings, time of day and lighting. If the two conflict on the structure itself, the STRUCTURE LOOK photo wins; for everything else, the LOCATION PHOTOGRAPH wins.

CAMERA & FRAME (follow exactly — this is the composition authority for this still; the reference images supply identity and place, never the framing):
- CAMERA: At the cat's eye height, hold a wide low-angle view from slightly behind and beside 검은 고양이, keeping its head turn prominent while the rooftop-room exterior and window remain visible ahead. Place the cat in the lower side of frame and the lit window higher across its line of sight, preserving lateral space toward the wall for the next tracking move.
- FRAMING SCALE: wide shot
- FRAME LAYOUT: 검은 고양이 in the lower-left of the frame, foreground, looks toward lit rooftop-room window; lit rooftop-room window in the upper-right of the frame, background.
- KEY BACKGROUND ELEMENTS: rooftop-room window (The room beyond is lit) — The window faces toward the cat and camera; illumination from the room is visible through it; used as Provides the elevated target of the cat's backward glance; rooftop-room exterior (Visible at dawn) — The window-bearing side faces the camera obliquely; used as Retains the location around the cat without overwhelming its head turn; wall (In the cat's path) — Its side extends laterally away from the window; used as Marks the direction the cat will take in the following movement.
Compose the frame exactly as specified above — camera angle, subject scale and screen placement. Keep true physical scale between people and background elements: fixtures and distant objects occupy only the small screen area their real size and distance dictate; never enlarge a background object into a foreground presence.

REALIZE FIGURATIVE LANGUAGE AS A LIVE-ACTION SHOT: the Korean shot
text may describe characters metaphorically, figuratively or with
exaggeration. Photograph what a real movie camera would actually
record on a physical set — exaggerated or figurative impressions
become realistic staging choices (lighting, distance, angle,
wardrobe), not literal fantasy imagery.
EVERY CHARACTER IS A HUMAN BEING: unless the story explicitly
features non-human or virtual beings (as in science fiction or
fantasy), every character — however indirectly, vaguely or
figuratively the text describes them — IS a real human. When the
text gives no direct visual description of a person, IMAGINE one
and still show them as a concrete, fully-formed human being:
always render the human form to the maximum extent — build,
posture, face, hands, clothing — never reduce a person to a shape,
blob, solid silhouette or abstract mass.

EXPRESSIONS ARE ACTED, NEVER ANATOMICAL: when the text describes a
person's eyes, face or presence emotionally or figuratively (vacant,
hollow, dazed, burning, lifeless gaze and the like), it describes an
ACTOR'S PERFORMANCE captured by a real camera — realize it ONLY
through gaze direction, focus, eyelids, facial muscles, stillness
and posture. Every living person's eyes remain anatomically normal
human eyes with a natural iris and pupil, natural sclera and normal
proportions; NEVER whiten, blank out, cloud over, glow, enlarge or
otherwise alter eyeballs, skin or anatomy — unless the story
explicitly declares that being non-human or supernatural in form.

PROPS FACE THE RIGHT WAY: every handheld or used object must be
oriented exactly as its real-world use requires. A person reading,
watching or operating something (a phone, a photograph, a paper,
any device) has its functional side — screen, front, page — facing
THEIR OWN eyes; the camera then sees whatever side the staging
geometry implies (often its back). Show the functional side to the
camera ONLY when the shot text itself stages it toward the viewer.
Never flip, mirror or reverse an object's front and back.

CARRIED STATE (persist exactly — must match the neighbouring shots of this scene): The rooftop room’s light remains on as the black cat looks toward it.

NO PEOPLE IN THIS SHOT: the shot text shows only the place and its state — no living person or any body part appears in frame, unless the shot text itself explicitly says so.

No text, captions, watermarks or annotations anywhere.

Create ONE FINAL photorealistic live-action film still of the moment below — contemporary South Korea, 2026; all people are Korean unless stated. TIME OF DAY (lock): dawn.

SHOT TEXT (authoritative, Korean): 새벽의 푸르스름한 빛이 감도는 옥탑방 외벽을 배경으로 검은 고양이가 고개를 돌려 창문을 바라보는 모습

LOCATION (lock): Outside on the villa rooftop at dawn, beside the rooftop apartment's wall and its illuminated window. The shot takes place here — the attached LOCATION PHOTOGRAPH shows the exact spot.

STRUCTURE LOOK AUTHORITY: the attached STRUCTURE LOOK photograph is the identity of the fixed structure at this location — wherever that structure appears in the frame, its shape, proportions, openings, materials and colors are LOCKED to it. The LOCATION PHOTOGRAPH remains the authority for this shot's sub-space, surroundings, time of day and lighting. If the two conflict on the structure itself, the STRUCTURE LOOK photo wins; for everything else, the LOCATION PHOTOGRAPH wins.

CAMERA & FRAME (follow exactly — this is the composition authority for this still; the reference images supply identity and place, never the framing):
- CAMERA: At the cat's eye height, hold a wide low-angle view from slightly behind and beside 검은 고양이, keeping its head turn prominent while the rooftop-room exterior and window remain visible ahead. Place the cat in the lower side of frame and the lit window higher across its line of sight, preserving lateral space toward the wall for the next tracking move.
- FRAMING SCALE: wide shot
- FRAME LAYOUT: 검은 고양이 in the lower-left of the frame, foreground, looks toward lit rooftop-room window; lit rooftop-room window in the upper-right of the frame, background.
- KEY BACKGROUND ELEMENTS: rooftop-room window (The room beyond is lit) — The window faces toward the cat and camera; illumination from the room is visible through it; used as Provides the elevated target of the cat's backward glance; rooftop-room exterior (Visible at dawn) — The window-bearing side faces the camera obliquely; used as Retains the location around the cat without overwhelming its head turn; wall (In the cat's path) — Its side extends laterally away from the window; used as Marks the direction the cat will take in the following movement.
Compose the frame exactly as specified above — camera angle, subject scale and screen placement. Keep true physical scale between people and background elements: fixtures and distant objects occupy only the small screen area their real size and distance dictate; never enlarge a background object into a foreground presence.

REALIZE FIGURATIVE LANGUAGE AS A LIVE-ACTION SHOT: the Korean shot
text may describe characters metaphorically, figuratively or with
exaggeration. Photograph what a real movie camera would actually
record on a physical set — exaggerated or figurative impressions
become realistic staging choices (lighting, distance, angle,
wardrobe), not literal fantasy imagery.
EVERY CHARACTER IS A HUMAN BEING: unless the story explicitly
features non-human or virtual beings (as in science fiction or
fantasy), every character — however indirectly, vaguely or
figuratively the text describes them — IS a real human. When the
text gives no direct visual description of a person, IMAGINE one
and still show them as a concrete, fully-formed human being:
always render the human form to the maximum extent — build,
posture, face, hands, clothing — never reduce a person to a shape,
blob, solid silhouette or abstract mass.

EXPRESSIONS ARE ACTED, NEVER ANATOMICAL: when the text describes a
person's eyes, face or presence emotionally or figuratively (vacant,
hollow, dazed, burning, lifeless gaze and the like), it describes an
ACTOR'S PERFORMANCE captured by a real camera — realize it ONLY
through gaze direction, focus, eyelids, facial muscles, stillness
and posture. Every living person's eyes remain anatomically normal
human eyes with a natural iris and pupil, natural sclera and normal
proportions; NEVER whiten, blank out, cloud over, glow, enlarge or
otherwise alter eyeballs, skin or anatomy — unless the story
explicitly declares that being non-human or supernatural in form.

PROPS FACE THE RIGHT WAY: every handheld or used object must be
oriented exactly as its real-world use requires. A person reading,
watching or operating something (a phone, a photograph, a paper,
any device) has its functional side — screen, front, page — facing
THEIR OWN eyes; the camera then sees whatever side the staging
geometry implies (often its back). Show the functional side to the
camera ONLY when the shot text itself stages it toward the viewer.
Never flip, mirror or reverse an object's front and back.

CARRIED STATE (persist exactly — must match the neighbouring shots of this scene): The rooftop room’s light remains on as the black cat looks toward it.

NO PEOPLE IN THIS SHOT: the shot text shows only the place and its state — no living person or any body part appears in frame, unless the shot text itself explicitly says so.

No text, captions, watermarks or annotations anywhere.

VISUAL APPROACH — this take (camera, framing, composition):
Use a broad, rectilinear lens feel with restrained perspective distortion and deep focus across the foreground and architecture. Build the composition from strong wall lines that carry the eye diagonally toward the elevated window, while preserving clean negative space along the lateral direction of movement. Let the cool dawn ambience model surfaces softly, with the window illumination held as a controlled practical glow and the cat’s turned head clearly separated by subtle edge contrast.

PRIORITY OF INSTRUCTIONS: the VISUAL APPROACH above governs camera,
framing and composition ONLY. Every other instruction in this prompt —
location, time of day, people and their identity, poses, props,
carried state, physical rules, and the no-text rule — is binding
exactly as written earlier. If the VISUAL APPROACH appears to conflict
with any of them, the earlier instruction wins.

Create ONE FINAL photorealistic live-action film still of the moment below — contemporary South Korea, 2026; all people are Korean unless stated. TIME OF DAY (lock): dawn.

SHOT TEXT (authoritative, Korean): 새벽의 푸르스름한 빛이 감도는 옥탑방 외벽을 배경으로 검은 고양이가 고개를 돌려 창문을 바라보는 모습

LOCATION (lock): Outside on the villa rooftop at dawn, beside the rooftop apartment's wall and its illuminated window. The shot takes place here — the attached LOCATION PHOTOGRAPH shows the exact spot.

STRUCTURE LOOK AUTHORITY: the attached STRUCTURE LOOK photograph is the identity of the fixed structure at this location — wherever that structure appears in the frame, its shape, proportions, openings, materials and colors are LOCKED to it. The LOCATION PHOTOGRAPH remains the authority for this shot's sub-space, surroundings, time of day and lighting. If the two conflict on the structure itself, the STRUCTURE LOOK photo wins; for everything else, the LOCATION PHOTOGRAPH wins.

CAMERA & FRAME (follow exactly — this is the composition authority for this still; the reference images supply identity and place, never the framing):
- CAMERA: At the cat's eye height, hold a wide low-angle view from slightly behind and beside 검은 고양이, keeping its head turn prominent while the rooftop-room exterior and window remain visible ahead. Place the cat in the lower side of frame and the lit window higher across its line of sight, preserving lateral space toward the wall for the next tracking move.
- FRAMING SCALE: wide shot
- FRAME LAYOUT: 검은 고양이 in the lower-left of the frame, foreground, looks toward lit rooftop-room window; lit rooftop-room window in the upper-right of the frame, background.
- KEY BACKGROUND ELEMENTS: rooftop-room window (The room beyond is lit) — The window faces toward the cat and camera; illumination from the room is visible through it; used as Provides the elevated target of the cat's backward glance; rooftop-room exterior (Visible at dawn) — The window-bearing side faces the camera obliquely; used as Retains the location around the cat without overwhelming its head turn; wall (In the cat's path) — Its side extends laterally away from the window; used as Marks the direction the cat will take in the following movement.
Compose the frame exactly as specified above — camera angle, subject scale and screen placement. Keep true physical scale between people and background elements: fixtures and distant objects occupy only the small screen area their real size and distance dictate; never enlarge a background object into a foreground presence.

REALIZE FIGURATIVE LANGUAGE AS A LIVE-ACTION SHOT: the Korean shot
text may describe characters metaphorically, figuratively or with
exaggeration. Photograph what a real movie camera would actually
record on a physical set — exaggerated or figurative impressions
become realistic staging choices (lighting, distance, angle,
wardrobe), not literal fantasy imagery.
EVERY CHARACTER IS A HUMAN BEING: unless the story explicitly
features non-human or virtual beings (as in science fiction or
fantasy), every character — however indirectly, vaguely or
figuratively the text describes them — IS a real human. When the
text gives no direct visual description of a person, IMAGINE one
and still show them as a concrete, fully-formed human being:
always render the human form to the maximum extent — build,
posture, face, hands, clothing — never reduce a person to a shape,
blob, solid silhouette or abstract mass.

EXPRESSIONS ARE ACTED, NEVER ANATOMICAL: when the text describes a
person's eyes, face or presence emotionally or figuratively (vacant,
hollow, dazed, burning, lifeless gaze and the like), it describes an
ACTOR'S PERFORMANCE captured by a real camera — realize it ONLY
through gaze direction, focus, eyelids, facial muscles, stillness
and posture. Every living person's eyes remain anatomically normal
human eyes with a natural iris and pupil, natural sclera and normal
proportions; NEVER whiten, blank out, cloud over, glow, enlarge or
otherwise alter eyeballs, skin or anatomy — unless the story
explicitly declares that being non-human or supernatural in form.

PROPS FACE THE RIGHT WAY: every handheld or used object must be
oriented exactly as its real-world use requires. A person reading,
watching or operating something (a phone, a photograph, a paper,
any device) has its functional side — screen, front, page — facing
THEIR OWN eyes; the camera then sees whatever side the staging
geometry implies (often its back). Show the functional side to the
camera ONLY when the shot text itself stages it toward the viewer.
Never flip, mirror or reverse an object's front and back.

CARRIED STATE (persist exactly — must match the neighbouring shots of this scene): The rooftop room’s light remains on as the black cat looks toward it.

NO PEOPLE IN THIS SHOT: the shot text shows only the place and its state — no living person or any body part appears in frame, unless the shot text itself explicitly says so.

No text, captions, watermarks or annotations anywhere.

VISUAL APPROACH — this take (camera, framing, composition):
Photograph with a gently compressed lens feel, stacking the foreground subject and distant façade into a tighter visual relationship while retaining the prescribed wide framing. Keep the strongest focus and micro-contrast on the turned head, allowing the exterior to fall slightly softer while the illuminated window remains unmistakable as a luminous destination. Use low dawn light for subdued tonal separation, with a restrained, contemplative contrast rather than graphic architectural sharpness.

PRIORITY OF INSTRUCTIONS: the VISUAL APPROACH above governs camera,
framing and composition ONLY. Every other instruction in this prompt —
location, time of day, people and their identity, poses, props,
carried state, physical rules, and the no-text rule — is binding
exactly as written earlier. If the VISUAL APPROACH appears to conflict
with any of them, the earlier instruction wins.

Edit this photograph. Apply ONLY the corrections listed below,
each in place, changing nothing else:

CORRECTIONS:
- Move the illuminated window from the right side wall to the front wall, positioning it to the left of the door to match the reference structure.
- Scale down the black cat significantly so it occupies only a small portion of the lower-left foreground to achieve a proper wide shot.

PRESERVE EVERYTHING ELSE EXACTLY as in the original photograph —
every person, pose, face, garment, object, set feature, lighting,
camera and framing stay identical. No text, captions or
watermarks anywhere.

Create ONE FINAL photorealistic live-action film still of the moment below — contemporary South Korea, 2026; all people are Korean unless stated. TIME OF DAY (lock): early morning.

SHOT TEXT (authoritative, Korean): 거울에 반사된 욕실 타일 벽면에 선명한 붉은 원 형태가 나타난 모습

LOCATION (lock): Inside the rooftop apartment bathroom, focused on the tiled wall reflected in the mirror in daytime. The shot takes place here — the attached LOCATION PHOTOGRAPH shows the exact spot.

CAMERA & FRAME (follow exactly — this is the composition authority for this still; the reference images supply identity and place, never the framing):
- CAMERA: Advance at eye level along the mirror's central axis until the reflected bathroom wall and its sharp red circle become the close central subject, cropping 수리영's reflection out at the end of the push. Preserve a narrow portion of the mirror boundary so the image is unmistakably a reflection rather than the actual wall.
- FRAMING SCALE: close-up
- FRAME LAYOUT: red circle reflected in the mirror in the middle-center of the frame, midground.
- KEY BACKGROUND ELEMENTS: bathroom mirror (Showing the red circle in reflection) — The mirror reflects the bathroom tile wall and the suddenly appearing red circle; a narrow edge of the mirror remains visible; used as Carries the apparition and defines the shot's reflected mode of perception; bathroom tile wall (A vivid red circle appears on its mirror image) — Its reflected face lies nearly square to the mirror axis; used as Appears only as reflected background behind the red mark.
Compose the frame exactly as specified above — camera angle, subject scale and screen placement. Keep true physical scale between people and background elements: fixtures and distant objects occupy only the small screen area their real size and distance dictate; never enlarge a background object into a foreground presence.

REALIZE FIGURATIVE LANGUAGE AS A LIVE-ACTION SHOT: the Korean shot
text may describe characters metaphorically, figuratively or with
exaggeration. Photograph what a real movie camera would actually
record on a physical set — exaggerated or figurative impressions
become realistic staging choices (lighting, distance, angle,
wardrobe), not literal fantasy imagery.
EVERY CHARACTER IS A HUMAN BEING: unless the story explicitly
features non-human or virtual beings (as in science fiction or
fantasy), every character — however indirectly, vaguely or
figuratively the text describes them — IS a real human. When the
text gives no direct visual description of a person, IMAGINE one
and still show them as a concrete, fully-formed human being:
always render the human form to the maximum extent — build,
posture, face, hands, clothing — never reduce a person to a shape,
blob, solid silhouette or abstract mass.

EXPRESSIONS ARE ACTED, NEVER ANATOMICAL: when the text describes a
person's eyes, face or presence emotionally or figuratively (vacant,
hollow, dazed, burning, lifeless gaze and the like), it describes an
ACTOR'S PERFORMANCE captured by a real camera — realize it ONLY
through gaze direction, focus, eyelids, facial muscles, stillness
and posture. Every living person's eyes remain anatomically normal
human eyes with a natural iris and pupil, natural sclera and normal
proportions; NEVER whiten, blank out, cloud over, glow, enlarge or
otherwise alter eyeballs, skin or anatomy — unless the story
explicitly declares that being non-human or supernatural in form.

PROPS FACE THE RIGHT WAY: every handheld or used object must be
oriented exactly as its real-world use requires. A person reading,
watching or operating something (a phone, a photograph, a paper,
any device) has its functional side — screen, front, page — facing
THEIR OWN eyes; the camera then sees whatever side the staging
geometry implies (often its back). Show the functional side to the
camera ONLY when the shot text itself stages it toward the viewer.
Never flip, mirror or reverse an object's front and back.

CARRIED STATE (persist exactly — must match the neighbouring shots of this scene): The home remains unnaturally clean, with Min-suk’s body and all blood still absent. A red circle is temporarily visible on the reflected bathroom wall until it vanishes when Su-ri-young turns toward it.

NO PEOPLE IN THIS SHOT: the shot text shows only the place and its state — no living person or any body part appears in frame, unless the shot text itself explicitly says so.

No text, captions, watermarks or annotations anywhere.

Create ONE FINAL photorealistic live-action film still of the moment below — contemporary South Korea, 2026; all people are Korean unless stated. TIME OF DAY (lock): early morning.

SHOT TEXT (authoritative, Korean): 거울에 반사된 욕실 타일 벽면에 선명한 붉은 원 형태가 나타난 모습

LOCATION (lock): Inside the rooftop apartment bathroom, focused on the tiled wall reflected in the mirror in daytime. The shot takes place here — the attached LOCATION PHOTOGRAPH shows the exact spot.

CAMERA & FRAME (follow exactly — this is the composition authority for this still; the reference images supply identity and place, never the framing):
- CAMERA: Advance at eye level along the mirror's central axis until the reflected bathroom wall and its sharp red circle become the close central subject, cropping 수리영's reflection out at the end of the push. Preserve a narrow portion of the mirror boundary so the image is unmistakably a reflection rather than the actual wall.
- FRAMING SCALE: close-up
- FRAME LAYOUT: red circle reflected in the mirror in the middle-center of the frame, midground.
- KEY BACKGROUND ELEMENTS: bathroom mirror (Showing the red circle in reflection) — The mirror reflects the bathroom tile wall and the suddenly appearing red circle; a narrow edge of the mirror remains visible; used as Carries the apparition and defines the shot's reflected mode of perception; bathroom tile wall (A vivid red circle appears on its mirror image) — Its reflected face lies nearly square to the mirror axis; used as Appears only as reflected background behind the red mark.
Compose the frame exactly as specified above — camera angle, subject scale and screen placement. Keep true physical scale between people and background elements: fixtures and distant objects occupy only the small screen area their real size and distance dictate; never enlarge a background object into a foreground presence.

REALIZE FIGURATIVE LANGUAGE AS A LIVE-ACTION SHOT: the Korean shot
text may describe characters metaphorically, figuratively or with
exaggeration. Photograph what a real movie camera would actually
record on a physical set — exaggerated or figurative impressions
become realistic staging choices (lighting, distance, angle,
wardrobe), not literal fantasy imagery.
EVERY CHARACTER IS A HUMAN BEING: unless the story explicitly
features non-human or virtual beings (as in science fiction or
fantasy), every character — however indirectly, vaguely or
figuratively the text describes them — IS a real human. When the
text gives no direct visual description of a person, IMAGINE one
and still show them as a concrete, fully-formed human being:
always render the human form to the maximum extent — build,
posture, face, hands, clothing — never reduce a person to a shape,
blob, solid silhouette or abstract mass.

EXPRESSIONS ARE ACTED, NEVER ANATOMICAL: when the text describes a
person's eyes, face or presence emotionally or figuratively (vacant,
hollow, dazed, burning, lifeless gaze and the like), it describes an
ACTOR'S PERFORMANCE captured by a real camera — realize it ONLY
through gaze direction, focus, eyelids, facial muscles, stillness
and posture. Every living person's eyes remain anatomically normal
human eyes with a natural iris and pupil, natural sclera and normal
proportions; NEVER whiten, blank out, cloud over, glow, enlarge or
otherwise alter eyeballs, skin or anatomy — unless the story
explicitly declares that being non-human or supernatural in form.

PROPS FACE THE RIGHT WAY: every handheld or used object must be
oriented exactly as its real-world use requires. A person reading,
watching or operating something (a phone, a photograph, a paper,
any device) has its functional side — screen, front, page — facing
THEIR OWN eyes; the camera then sees whatever side the staging
geometry implies (often its back). Show the functional side to the
camera ONLY when the shot text itself stages it toward the viewer.
Never flip, mirror or reverse an object's front and back.

CARRIED STATE (persist exactly — must match the neighbouring shots of this scene): The home remains unnaturally clean, with Min-suk’s body and all blood still absent. A red circle is temporarily visible on the reflected bathroom wall until it vanishes when Su-ri-young turns toward it.

NO PEOPLE IN THIS SHOT: the shot text shows only the place and its state — no living person or any body part appears in frame, unless the shot text itself explicitly says so.

No text, captions, watermarks or annotations anywhere.

VISUAL APPROACH — this take (camera, framing, composition):
Photograph the reflection as a clinically precise, head-on close-up with a restrained normal-lens feel and deep, even focus. Keep the tile grid geometrically rigid around the middle-centered red circle, with the narrow mirror boundary crisply legible. Use soft, cool early-morning illumination and controlled tonal separation, making the vivid red mark feel starkly physical against the unnaturally clean bathroom.

PRIORITY OF INSTRUCTIONS: the VISUAL APPROACH above governs camera,
framing and composition ONLY. Every other instruction in this prompt —
location, time of day, people and their identity, poses, props,
carried state, physical rules, and the no-text rule — is binding
exactly as written earlier. If the VISUAL APPROACH appears to conflict
with any of them, the earlier instruction wins.

Create ONE FINAL photorealistic live-action film still of the moment below — contemporary South Korea, 2026; all people are Korean unless stated. TIME OF DAY (lock): early morning.

SHOT TEXT (authoritative, Korean): 거울에 반사된 욕실 타일 벽면에 선명한 붉은 원 형태가 나타난 모습

LOCATION (lock): Inside the rooftop apartment bathroom, focused on the tiled wall reflected in the mirror in daytime. The shot takes place here — the attached LOCATION PHOTOGRAPH shows the exact spot.

CAMERA & FRAME (follow exactly — this is the composition authority for this still; the reference images supply identity and place, never the framing):
- CAMERA: Advance at eye level along the mirror's central axis until the reflected bathroom wall and its sharp red circle become the close central subject, cropping 수리영's reflection out at the end of the push. Preserve a narrow portion of the mirror boundary so the image is unmistakably a reflection rather than the actual wall.
- FRAMING SCALE: close-up
- FRAME LAYOUT: red circle reflected in the mirror in the middle-center of the frame, midground.
- KEY BACKGROUND ELEMENTS: bathroom mirror (Showing the red circle in reflection) — The mirror reflects the bathroom tile wall and the suddenly appearing red circle; a narrow edge of the mirror remains visible; used as Carries the apparition and defines the shot's reflected mode of perception; bathroom tile wall (A vivid red circle appears on its mirror image) — Its reflected face lies nearly square to the mirror axis; used as Appears only as reflected background behind the red mark.
Compose the frame exactly as specified above — camera angle, subject scale and screen placement. Keep true physical scale between people and background elements: fixtures and distant objects occupy only the small screen area their real size and distance dictate; never enlarge a background object into a foreground presence.

REALIZE FIGURATIVE LANGUAGE AS A LIVE-ACTION SHOT: the Korean shot
text may describe characters metaphorically, figuratively or with
exaggeration. Photograph what a real movie camera would actually
record on a physical set — exaggerated or figurative impressions
become realistic staging choices (lighting, distance, angle,
wardrobe), not literal fantasy imagery.
EVERY CHARACTER IS A HUMAN BEING: unless the story explicitly
features non-human or virtual beings (as in science fiction or
fantasy), every character — however indirectly, vaguely or
figuratively the text describes them — IS a real human. When the
text gives no direct visual description of a person, IMAGINE one
and still show them as a concrete, fully-formed human being:
always render the human form to the maximum extent — build,
posture, face, hands, clothing — never reduce a person to a shape,
blob, solid silhouette or abstract mass.

EXPRESSIONS ARE ACTED, NEVER ANATOMICAL: when the text describes a
person's eyes, face or presence emotionally or figuratively (vacant,
hollow, dazed, burning, lifeless gaze and the like), it describes an
ACTOR'S PERFORMANCE captured by a real camera — realize it ONLY
through gaze direction, focus, eyelids, facial muscles, stillness
and posture. Every living person's eyes remain anatomically normal
human eyes with a natural iris and pupil, natural sclera and normal
proportions; NEVER whiten, blank out, cloud over, glow, enlarge or
otherwise alter eyeballs, skin or anatomy — unless the story
explicitly declares that being non-human or supernatural in form.

PROPS FACE THE RIGHT WAY: every handheld or used object must be
oriented exactly as its real-world use requires. A person reading,
watching or operating something (a phone, a photograph, a paper,
any device) has its functional side — screen, front, page — facing
THEIR OWN eyes; the camera then sees whatever side the staging
geometry implies (often its back). Show the functional side to the
camera ONLY when the shot text itself stages it toward the viewer.
Never flip, mirror or reverse an object's front and back.

CARRIED STATE (persist exactly — must match the neighbouring shots of this scene): The home remains unnaturally clean, with Min-suk’s body and all blood still absent. A red circle is temporarily visible on the reflected bathroom wall until it vanishes when Su-ri-young turns toward it.

NO PEOPLE IN THIS SHOT: the shot text shows only the place and its state — no living person or any body part appears in frame, unless the shot text itself explicitly says so.

No text, captions, watermarks or annotations anywhere.

VISUAL APPROACH — this take (camera, framing, composition):
Use a compressed long-lens feel from eye level, isolating the reflected red circle in a tight plane of sharp focus while the retained mirror boundary falls into gentle foreground softness. Let directional early-morning light graze the reflected tiles, revealing faint surface texture and subdued tonal falloff around the sharply defined mark. The composition should feel hushed and ominously observational, with the red circle commanding the center without appearing luminous.

PRIORITY OF INSTRUCTIONS: the VISUAL APPROACH above governs camera,
framing and composition ONLY. Every other instruction in this prompt —
location, time of day, people and their identity, poses, props,
carried state, physical rules, and the no-text rule — is binding
exactly as written earlier. If the VISUAL APPROACH appears to conflict
with any of them, the earlier instruction wins.

Edit this photograph. Apply ONLY the corrections listed below,
each in place, changing nothing else:

CORRECTIONS:
- Remove the glass shower door and window from the reflection, replacing them with plain white tiles that match the reference bathroom.

PRESERVE EVERYTHING ELSE EXACTLY as in the original photograph —
every person, pose, face, garment, object, set feature, lighting,
camera and framing stay identical. No text, captions or
watermarks anywhere.

Create ONE FINAL photorealistic live-action film still of the moment below — contemporary South Korea, 2026; all people are Korean unless stated. TIME OF DAY (lock): early morning.

SHOT TEXT (authoritative, Korean): 액자 속 강민숙의 얼굴 부분이 형체를 알아볼 수 없게 기괴하게 뭉개져 있는 클로즈업

LOCATION (lock): Inside the rooftop apartment's living and dining area, at the table where a framed family photograph is displayed in daylight. The shot takes place here — the attached LOCATION PHOTOGRAPH shows the exact spot.

CAMERA & FRAME (follow exactly — this is the composition authority for this still; the reference images supply identity and place, never the framing):
- CAMERA: From chest height beside the living-room table, complete the dolly-in until the framed photograph is close and 강민숙 (사진 속 얼굴 왜곡 상태)'s unrecognizable face occupies most of the composition without excluding the photograph's edge. Keep the picture surface nearly square to camera so the grotesque downward deformation reads as altered content within the photograph, not damage to the viewing lens.
- FRAMING SCALE: close-up
- KEY BACKGROUND ELEMENTS: framed photograph (강민숙's face appears grotesquely distorted while the photograph is viewed) — Its image-bearing front faces the camera, showing 강민숙's smiling portrait transformed into an unrecognizable, downward-smeared face; used as Acts as the primary content surface on which the facial distortion manifests; living-room table (Holding the framed photograph) — A narrow portion of its upper side is visible beneath the photograph; used as Provides a small lower-edge reference establishing the photograph's placement in the room.
Compose the frame exactly as specified above — camera angle, subject scale and screen placement. Keep true physical scale between people and background elements: fixtures and distant objects occupy only the small screen area their real size and distance dictate; never enlarge a background object into a foreground presence.

REALIZE FIGURATIVE LANGUAGE AS A LIVE-ACTION SHOT: the Korean shot
text may describe characters metaphorically, figuratively or with
exaggeration. Photograph what a real movie camera would actually
record on a physical set — exaggerated or figurative impressions
become realistic staging choices (lighting, distance, angle,
wardrobe), not literal fantasy imagery.
EVERY CHARACTER IS A HUMAN BEING: unless the story explicitly
features non-human or virtual beings (as in science fiction or
fantasy), every character — however indirectly, vaguely or
figuratively the text describes them — IS a real human. When the
text gives no direct visual description of a person, IMAGINE one
and still show them as a concrete, fully-formed human being:
always render the human form to the maximum extent — build,
posture, face, hands, clothing — never reduce a person to a shape,
blob, solid silhouette or abstract mass.

EXPRESSIONS ARE ACTED, NEVER ANATOMICAL: when the text describes a
person's eyes, face or presence emotionally or figuratively (vacant,
hollow, dazed, burning, lifeless gaze and the like), it describes an
ACTOR'S PERFORMANCE captured by a real camera — realize it ONLY
through gaze direction, focus, eyelids, facial muscles, stillness
and posture. Every living person's eyes remain anatomically normal
human eyes with a natural iris and pupil, natural sclera and normal
proportions; NEVER whiten, blank out, cloud over, glow, enlarge or
otherwise alter eyeballs, skin or anatomy — unless the story
explicitly declares that being non-human or supernatural in form.

PROPS FACE THE RIGHT WAY: every handheld or used object must be
oriented exactly as its real-world use requires. A person reading,
watching or operating something (a phone, a photograph, a paper,
any device) has its functional side — screen, front, page — facing
THEIR OWN eyes; the camera then sees whatever side the staging
geometry implies (often its back). Show the functional side to the
camera ONLY when the shot text itself stages it toward the viewer.
Never flip, mirror or reverse an object's front and back.

CARRIED STATE (persist exactly — must match the neighbouring shots of this scene): The framed photograph remains on the table; Min-suk’s face appears grotesquely melted in this moment but returns to normal when Su-ri-young looks again. The home otherwise remains clean and corpse-free.

NO PEOPLE IN THIS SHOT: the shot text shows only the place and its state — no living person or any body part appears in frame, unless the shot text itself explicitly says so.

No text, captions, watermarks or annotations anywhere.

Create ONE FINAL photorealistic live-action film still of the moment below — contemporary South Korea, 2026; all people are Korean unless stated. TIME OF DAY (lock): early morning.

SHOT TEXT (authoritative, Korean): 액자 속 강민숙의 얼굴 부분이 형체를 알아볼 수 없게 기괴하게 뭉개져 있는 클로즈업

LOCATION (lock): Inside the rooftop apartment's living and dining area, at the table where a framed family photograph is displayed in daylight. The shot takes place here — the attached LOCATION PHOTOGRAPH shows the exact spot.

CAMERA & FRAME (follow exactly — this is the composition authority for this still; the reference images supply identity and place, never the framing):
- CAMERA: From chest height beside the living-room table, complete the dolly-in until the framed photograph is close and 강민숙 (사진 속 얼굴 왜곡 상태)'s unrecognizable face occupies most of the composition without excluding the photograph's edge. Keep the picture surface nearly square to camera so the grotesque downward deformation reads as altered content within the photograph, not damage to the viewing lens.
- FRAMING SCALE: close-up
- KEY BACKGROUND ELEMENTS: framed photograph (강민숙's face appears grotesquely distorted while the photograph is viewed) — Its image-bearing front faces the camera, showing 강민숙's smiling portrait transformed into an unrecognizable, downward-smeared face; used as Acts as the primary content surface on which the facial distortion manifests; living-room table (Holding the framed photograph) — A narrow portion of its upper side is visible beneath the photograph; used as Provides a small lower-edge reference establishing the photograph's placement in the room.
Compose the frame exactly as specified above — camera angle, subject scale and screen placement. Keep true physical scale between people and background elements: fixtures and distant objects occupy only the small screen area their real size and distance dictate; never enlarge a background object into a foreground presence.

REALIZE FIGURATIVE LANGUAGE AS A LIVE-ACTION SHOT: the Korean shot
text may describe characters metaphorically, figuratively or with
exaggeration. Photograph what a real movie camera would actually
record on a physical set — exaggerated or figurative impressions
become realistic staging choices (lighting, distance, angle,
wardrobe), not literal fantasy imagery.
EVERY CHARACTER IS A HUMAN BEING: unless the story explicitly
features non-human or virtual beings (as in science fiction or
fantasy), every character — however indirectly, vaguely or
figuratively the text describes them — IS a real human. When the
text gives no direct visual description of a person, IMAGINE one
and still show them as a concrete, fully-formed human being:
always render the human form to the maximum extent — build,
posture, face, hands, clothing — never reduce a person to a shape,
blob, solid silhouette or abstract mass.

EXPRESSIONS ARE ACTED, NEVER ANATOMICAL: when the text describes a
person's eyes, face or presence emotionally or figuratively (vacant,
hollow, dazed, burning, lifeless gaze and the like), it describes an
ACTOR'S PERFORMANCE captured by a real camera — realize it ONLY
through gaze direction, focus, eyelids, facial muscles, stillness
and posture. Every living person's eyes remain anatomically normal
human eyes with a natural iris and pupil, natural sclera and normal
proportions; NEVER whiten, blank out, cloud over, glow, enlarge or
otherwise alter eyeballs, skin or anatomy — unless the story
explicitly declares that being non-human or supernatural in form.

PROPS FACE THE RIGHT WAY: every handheld or used object must be
oriented exactly as its real-world use requires. A person reading,
watching or operating something (a phone, a photograph, a paper,
any device) has its functional side — screen, front, page — facing
THEIR OWN eyes; the camera then sees whatever side the staging
geometry implies (often its back). Show the functional side to the
camera ONLY when the shot text itself stages it toward the viewer.
Never flip, mirror or reverse an object's front and back.

CARRIED STATE (persist exactly — must match the neighbouring shots of this scene): The framed photograph remains on the table; Min-suk’s face appears grotesquely melted in this moment but returns to normal when Su-ri-young looks again. The home otherwise remains clean and corpse-free.

NO PEOPLE IN THIS SHOT: the shot text shows only the place and its state — no living person or any body part appears in frame, unless the shot text itself explicitly says so.

No text, captions, watermarks or annotations anywhere.

VISUAL APPROACH — this take (camera, framing, composition):
Photograph the close-up with rigid frontal geometry and unnerving forensic clarity. Use a restrained, natural perspective and enough depth of field to hold the deformed facial image, the frame edge, and the narrow table reference cleanly legible. Soft, cool early-morning daylight falls evenly across the picture surface, making the impossible alteration feel starkly factual rather than stylized.

PRIORITY OF INSTRUCTIONS: the VISUAL APPROACH above governs camera,
framing and composition ONLY. Every other instruction in this prompt —
location, time of day, people and their identity, poses, props,
carried state, physical rules, and the no-text rule — is binding
exactly as written earlier. If the VISUAL APPROACH appears to conflict
with any of them, the earlier instruction wins.

Create ONE FINAL photorealistic live-action film still of the moment below — contemporary South Korea, 2026; all people are Korean unless stated. TIME OF DAY (lock): early morning.

SHOT TEXT (authoritative, Korean): 액자 속 강민숙의 얼굴 부분이 형체를 알아볼 수 없게 기괴하게 뭉개져 있는 클로즈업

LOCATION (lock): Inside the rooftop apartment's living and dining area, at the table where a framed family photograph is displayed in daylight. The shot takes place here — the attached LOCATION PHOTOGRAPH shows the exact spot.

CAMERA & FRAME (follow exactly — this is the composition authority for this still; the reference images supply identity and place, never the framing):
- CAMERA: From chest height beside the living-room table, complete the dolly-in until the framed photograph is close and 강민숙 (사진 속 얼굴 왜곡 상태)'s unrecognizable face occupies most of the composition without excluding the photograph's edge. Keep the picture surface nearly square to camera so the grotesque downward deformation reads as altered content within the photograph, not damage to the viewing lens.
- FRAMING SCALE: close-up
- KEY BACKGROUND ELEMENTS: framed photograph (강민숙's face appears grotesquely distorted while the photograph is viewed) — Its image-bearing front faces the camera, showing 강민숙's smiling portrait transformed into an unrecognizable, downward-smeared face; used as Acts as the primary content surface on which the facial distortion manifests; living-room table (Holding the framed photograph) — A narrow portion of its upper side is visible beneath the photograph; used as Provides a small lower-edge reference establishing the photograph's placement in the room.
Compose the frame exactly as specified above — camera angle, subject scale and screen placement. Keep true physical scale between people and background elements: fixtures and distant objects occupy only the small screen area their real size and distance dictate; never enlarge a background object into a foreground presence.

REALIZE FIGURATIVE LANGUAGE AS A LIVE-ACTION SHOT: the Korean shot
text may describe characters metaphorically, figuratively or with
exaggeration. Photograph what a real movie camera would actually
record on a physical set — exaggerated or figurative impressions
become realistic staging choices (lighting, distance, angle,
wardrobe), not literal fantasy imagery.
EVERY CHARACTER IS A HUMAN BEING: unless the story explicitly
features non-human or virtual beings (as in science fiction or
fantasy), every character — however indirectly, vaguely or
figuratively the text describes them — IS a real human. When the
text gives no direct visual description of a person, IMAGINE one
and still show them as a concrete, fully-formed human being:
always render the human form to the maximum extent — build,
posture, face, hands, clothing — never reduce a person to a shape,
blob, solid silhouette or abstract mass.

EXPRESSIONS ARE ACTED, NEVER ANATOMICAL: when the text describes a
person's eyes, face or presence emotionally or figuratively (vacant,
hollow, dazed, burning, lifeless gaze and the like), it describes an
ACTOR'S PERFORMANCE captured by a real camera — realize it ONLY
through gaze direction, focus, eyelids, facial muscles, stillness
and posture. Every living person's eyes remain anatomically normal
human eyes with a natural iris and pupil, natural sclera and normal
proportions; NEVER whiten, blank out, cloud over, glow, enlarge or
otherwise alter eyeballs, skin or anatomy — unless the story
explicitly declares that being non-human or supernatural in form.

PROPS FACE THE RIGHT WAY: every handheld or used object must be
oriented exactly as its real-world use requires. A person reading,
watching or operating something (a phone, a photograph, a paper,
any device) has its functional side — screen, front, page — facing
THEIR OWN eyes; the camera then sees whatever side the staging
geometry implies (often its back). Show the functional side to the
camera ONLY when the shot text itself stages it toward the viewer.
Never flip, mirror or reverse an object's front and back.

CARRIED STATE (persist exactly — must match the neighbouring shots of this scene): The framed photograph remains on the table; Min-suk’s face appears grotesquely melted in this moment but returns to normal when Su-ri-young looks again. The home otherwise remains clean and corpse-free.

NO PEOPLE IN THIS SHOT: the shot text shows only the place and its state — no living person or any body part appears in frame, unless the shot text itself explicitly says so.

No text, captions, watermarks or annotations anywhere.

VISUAL APPROACH — this take (camera, framing, composition):
Create a claustrophobic, subjective close-focus image, keeping the grotesquely dragged facial area critically sharp while the visible frame edge and lower table sliver fall into gentle focus softness. Let directional early-morning daylight skim across the nearly frontal picture surface, with restrained shadow shaping that draws the eye downward through the deformation. Preserve realistic photographic texture and subdued color, giving the instant a hushed, intimate dread.

PRIORITY OF INSTRUCTIONS: the VISUAL APPROACH above governs camera,
framing and composition ONLY. Every other instruction in this prompt —
location, time of day, people and their identity, poses, props,
carried state, physical rules, and the no-text rule — is binding
exactly as written earlier. If the VISUAL APPROACH appears to conflict
with any of them, the earlier instruction wins.

Edit this photograph. Apply ONLY the corrections listed below,
each in place, changing nothing else:

CORRECTIONS:
- Dolly the camera in significantly closer so that the distorted face occupies the majority of the composition, leaving only the photograph's edges and a narrow strip of the table visible.
- Confine the downward melting distortion entirely within the printed image area of the photograph, removing the spill that extends onto the physical white mat and silver frame.

PRESERVE EVERYTHING ELSE EXACTLY as in the original photograph —
every person, pose, face, garment, object, set feature, lighting,
camera and framing stay identical. No text, captions or
watermarks anywhere.

Create ONE FINAL photorealistic live-action film still of the moment below — contemporary South Korea, 2026; all people are Korean unless stated. TIME OF DAY (lock): late morning.

SHOT TEXT (authoritative, Korean): 모니터 화면에 해상 지도와 붉은 점, 좌표 숫자가 선명하게 떠 있는 클로즈업

LOCATION (lock): At a detective's computer workstation inside the daytime police office, where the phone-tracking result is displayed on the monitor. The shot takes place here — the attached LOCATION PHOTOGRAPH shows the exact spot.

CAMERA & FRAME (follow exactly — this is the composition authority for this still; the reference images supply identity and place, never the framing):
- CAMERA: Move directly toward the monitor and stop at the specified close viewing distance, keeping the bezel and a narrow band of the surrounding desk visible so the display remains below an exaggerated object scale. The maritime map, sharp crimson location point, coordinate figures, timestamp, and offline status are read through the device screen in a centered, clinically legible composition.
- FRAMING SCALE: close-up
- KEY BACKGROUND ELEMENTS: 김형사 모니터 (On with the location-tracking result open) — The front display faces the camera and shows 강민숙's last-location result: maritime map, marked point, coordinates 37.484400 N / 126.072486 E, timestamp 2026-02-18 11:21:08, GPS/Network source, and offline status; used as Primary device-mediated focal surface at the center of the frame; 경찰서 책상 (In use) — Its near edge remains visible below and beside the monitor; used as Provides environmental scale around the close monitor view; 살인사건 확대 사진들 (Scattered across the desk) — Their image-bearing fronts face upward on the desk, with only partial photographic content visible at frame edges; used as Peripheral evidence at the desk edge, kept subordinate to the location result.
Compose the frame exactly as specified above — camera angle, subject scale and screen placement. Keep true physical scale between people and background elements: fixtures and distant objects occupy only the small screen area their real size and distance dictate; never enlarge a background object into a foreground presence.

REALIZE FIGURATIVE LANGUAGE AS A LIVE-ACTION SHOT: the Korean shot
text may describe characters metaphorically, figuratively or with
exaggeration. Photograph what a real movie camera would actually
record on a physical set — exaggerated or figurative impressions
become realistic staging choices (lighting, distance, angle,
wardrobe), not literal fantasy imagery.
EVERY CHARACTER IS A HUMAN BEING: unless the story explicitly
features non-human or virtual beings (as in science fiction or
fantasy), every character — however indirectly, vaguely or
figuratively the text describes them — IS a real human. When the
text gives no direct visual description of a person, IMAGINE one
and still show them as a concrete, fully-formed human being:
always render the human form to the maximum extent — build,
posture, face, hands, clothing — never reduce a person to a shape,
blob, solid silhouette or abstract mass.

EXPRESSIONS ARE ACTED, NEVER ANATOMICAL: when the text describes a
person's eyes, face or presence emotionally or figuratively (vacant,
hollow, dazed, burning, lifeless gaze and the like), it describes an
ACTOR'S PERFORMANCE captured by a real camera — realize it ONLY
through gaze direction, focus, eyelids, facial muscles, stillness
and posture. Every living person's eyes remain anatomically normal
human eyes with a natural iris and pupil, natural sclera and normal
proportions; NEVER whiten, blank out, cloud over, glow, enlarge or
otherwise alter eyeballs, skin or anatomy — unless the story
explicitly declares that being non-human or supernatural in form.

PROPS FACE THE RIGHT WAY: every handheld or used object must be
oriented exactly as its real-world use requires. A person reading,
watching or operating something (a phone, a photograph, a paper,
any device) has its functional side — screen, front, page — facing
THEIR OWN eyes; the camera then sees whatever side the staging
geometry implies (often its back). Show the functional side to the
camera ONLY when the shot text itself stages it toward the viewer.
Never flip, mirror or reverse an object's front and back.

CARRIED STATE (persist exactly — must match the neighbouring shots of this scene): Kim’s monitor displays Min-suk’s last phone location as an offshore map coordinate, with the phone marked offline. The murder-scene photographs remain scattered across Hye-su’s desk.

NO PEOPLE IN THIS SHOT: the shot text shows only the place and its state — no living person or any body part appears in frame, unless the shot text itself explicitly says so.

No text, captions, watermarks or annotations anywhere.

Create ONE FINAL photorealistic live-action film still of the moment below — contemporary South Korea, 2026; all people are Korean unless stated. TIME OF DAY (lock): late morning.

SHOT TEXT (authoritative, Korean): 모니터 화면에 해상 지도와 붉은 점, 좌표 숫자가 선명하게 떠 있는 클로즈업

LOCATION (lock): At a detective's computer workstation inside the daytime police office, where the phone-tracking result is displayed on the monitor. The shot takes place here — the attached LOCATION PHOTOGRAPH shows the exact spot.

CAMERA & FRAME (follow exactly — this is the composition authority for this still; the reference images supply identity and place, never the framing):
- CAMERA: Move directly toward the monitor and stop at the specified close viewing distance, keeping the bezel and a narrow band of the surrounding desk visible so the display remains below an exaggerated object scale. The maritime map, sharp crimson location point, coordinate figures, timestamp, and offline status are read through the device screen in a centered, clinically legible composition.
- FRAMING SCALE: close-up
- KEY BACKGROUND ELEMENTS: 김형사 모니터 (On with the location-tracking result open) — The front display faces the camera and shows 강민숙's last-location result: maritime map, marked point, coordinates 37.484400 N / 126.072486 E, timestamp 2026-02-18 11:21:08, GPS/Network source, and offline status; used as Primary device-mediated focal surface at the center of the frame; 경찰서 책상 (In use) — Its near edge remains visible below and beside the monitor; used as Provides environmental scale around the close monitor view; 살인사건 확대 사진들 (Scattered across the desk) — Their image-bearing fronts face upward on the desk, with only partial photographic content visible at frame edges; used as Peripheral evidence at the desk edge, kept subordinate to the location result.
Compose the frame exactly as specified above — camera angle, subject scale and screen placement. Keep true physical scale between people and background elements: fixtures and distant objects occupy only the small screen area their real size and distance dictate; never enlarge a background object into a foreground presence.

REALIZE FIGURATIVE LANGUAGE AS A LIVE-ACTION SHOT: the Korean shot
text may describe characters metaphorically, figuratively or with
exaggeration. Photograph what a real movie camera would actually
record on a physical set — exaggerated or figurative impressions
become realistic staging choices (lighting, distance, angle,
wardrobe), not literal fantasy imagery.
EVERY CHARACTER IS A HUMAN BEING: unless the story explicitly
features non-human or virtual beings (as in science fiction or
fantasy), every character — however indirectly, vaguely or
figuratively the text describes them — IS a real human. When the
text gives no direct visual description of a person, IMAGINE one
and still show them as a concrete, fully-formed human being:
always render the human form to the maximum extent — build,
posture, face, hands, clothing — never reduce a person to a shape,
blob, solid silhouette or abstract mass.

EXPRESSIONS ARE ACTED, NEVER ANATOMICAL: when the text describes a
person's eyes, face or presence emotionally or figuratively (vacant,
hollow, dazed, burning, lifeless gaze and the like), it describes an
ACTOR'S PERFORMANCE captured by a real camera — realize it ONLY
through gaze direction, focus, eyelids, facial muscles, stillness
and posture. Every living person's eyes remain anatomically normal
human eyes with a natural iris and pupil, natural sclera and normal
proportions; NEVER whiten, blank out, cloud over, glow, enlarge or
otherwise alter eyeballs, skin or anatomy — unless the story
explicitly declares that being non-human or supernatural in form.

PROPS FACE THE RIGHT WAY: every handheld or used object must be
oriented exactly as its real-world use requires. A person reading,
watching or operating something (a phone, a photograph, a paper,
any device) has its functional side — screen, front, page — facing
THEIR OWN eyes; the camera then sees whatever side the staging
geometry implies (often its back). Show the functional side to the
camera ONLY when the shot text itself stages it toward the viewer.
Never flip, mirror or reverse an object's front and back.

CARRIED STATE (persist exactly — must match the neighbouring shots of this scene): Kim’s monitor displays Min-suk’s last phone location as an offshore map coordinate, with the phone marked offline. The murder-scene photographs remain scattered across Hye-su’s desk.

NO PEOPLE IN THIS SHOT: the shot text shows only the place and its state — no living person or any body part appears in frame, unless the shot text itself explicitly says so.

No text, captions, watermarks or annotations anywhere.

VISUAL APPROACH — this take (camera, framing, composition):
Photograph the monitor dead-on with the screen plane perfectly square to the camera, using a natural close-focus perspective and broad, even sharpness across the display, bezel, and narrow desk band. Keep exposure neutral and controlled under soft late-morning office ambience, suppressing glare so the tracking result reads with crisp forensic clarity while the peripheral evidence remains visually subordinate.

PRIORITY OF INSTRUCTIONS: the VISUAL APPROACH above governs camera,
framing and composition ONLY. Every other instruction in this prompt —
location, time of day, people and their identity, poses, props,
carried state, physical rules, and the no-text rule — is binding
exactly as written earlier. If the VISUAL APPROACH appears to conflict
with any of them, the earlier instruction wins.

Create ONE FINAL photorealistic live-action film still of the moment below — contemporary South Korea, 2026; all people are Korean unless stated. TIME OF DAY (lock): late morning.

SHOT TEXT (authoritative, Korean): 모니터 화면에 해상 지도와 붉은 점, 좌표 숫자가 선명하게 떠 있는 클로즈업

LOCATION (lock): At a detective's computer workstation inside the daytime police office, where the phone-tracking result is displayed on the monitor. The shot takes place here — the attached LOCATION PHOTOGRAPH shows the exact spot.

CAMERA & FRAME (follow exactly — this is the composition authority for this still; the reference images supply identity and place, never the framing):
- CAMERA: Move directly toward the monitor and stop at the specified close viewing distance, keeping the bezel and a narrow band of the surrounding desk visible so the display remains below an exaggerated object scale. The maritime map, sharp crimson location point, coordinate figures, timestamp, and offline status are read through the device screen in a centered, clinically legible composition.
- FRAMING SCALE: close-up
- KEY BACKGROUND ELEMENTS: 김형사 모니터 (On with the location-tracking result open) — The front display faces the camera and shows 강민숙's last-location result: maritime map, marked point, coordinates 37.484400 N / 126.072486 E, timestamp 2026-02-18 11:21:08, GPS/Network source, and offline status; used as Primary device-mediated focal surface at the center of the frame; 경찰서 책상 (In use) — Its near edge remains visible below and beside the monitor; used as Provides environmental scale around the close monitor view; 살인사건 확대 사진들 (Scattered across the desk) — Their image-bearing fronts face upward on the desk, with only partial photographic content visible at frame edges; used as Peripheral evidence at the desk edge, kept subordinate to the location result.
Compose the frame exactly as specified above — camera angle, subject scale and screen placement. Keep true physical scale between people and background elements: fixtures and distant objects occupy only the small screen area their real size and distance dictate; never enlarge a background object into a foreground presence.

REALIZE FIGURATIVE LANGUAGE AS A LIVE-ACTION SHOT: the Korean shot
text may describe characters metaphorically, figuratively or with
exaggeration. Photograph what a real movie camera would actually
record on a physical set — exaggerated or figurative impressions
become realistic staging choices (lighting, distance, angle,
wardrobe), not literal fantasy imagery.
EVERY CHARACTER IS A HUMAN BEING: unless the story explicitly
features non-human or virtual beings (as in science fiction or
fantasy), every character — however indirectly, vaguely or
figuratively the text describes them — IS a real human. When the
text gives no direct visual description of a person, IMAGINE one
and still show them as a concrete, fully-formed human being:
always render the human form to the maximum extent — build,
posture, face, hands, clothing — never reduce a person to a shape,
blob, solid silhouette or abstract mass.

EXPRESSIONS ARE ACTED, NEVER ANATOMICAL: when the text describes a
person's eyes, face or presence emotionally or figuratively (vacant,
hollow, dazed, burning, lifeless gaze and the like), it describes an
ACTOR'S PERFORMANCE captured by a real camera — realize it ONLY
through gaze direction, focus, eyelids, facial muscles, stillness
and posture. Every living person's eyes remain anatomically normal
human eyes with a natural iris and pupil, natural sclera and normal
proportions; NEVER whiten, blank out, cloud over, glow, enlarge or
otherwise alter eyeballs, skin or anatomy — unless the story
explicitly declares that being non-human or supernatural in form.

PROPS FACE THE RIGHT WAY: every handheld or used object must be
oriented exactly as its real-world use requires. A person reading,
watching or operating something (a phone, a photograph, a paper,
any device) has its functional side — screen, front, page — facing
THEIR OWN eyes; the camera then sees whatever side the staging
geometry implies (often its back). Show the functional side to the
camera ONLY when the shot text itself stages it toward the viewer.
Never flip, mirror or reverse an object's front and back.

CARRIED STATE (persist exactly — must match the neighbouring shots of this scene): Kim’s monitor displays Min-suk’s last phone location as an offshore map coordinate, with the phone marked offline. The murder-scene photographs remain scattered across Hye-su’s desk.

NO PEOPLE IN THIS SHOT: the shot text shows only the place and its state — no living person or any body part appears in frame, unless the shot text itself explicitly says so.

No text, captions, watermarks or annotations anywhere.

VISUAL APPROACH — this take (camera, framing, composition):
Hold exact focus on the device screen with a markedly shallow depth of field, allowing the bezel, desk edge, and partial photographs to fall rapidly into soft blur around the centered display. Let the monitor’s restrained luminance dominate the available late-morning office light, with gentle edge reflections and deeper surrounding tones creating a tense, isolated emphasis while every required screen detail remains sharply legible.

PRIORITY OF INSTRUCTIONS: the VISUAL APPROACH above governs camera,
framing and composition ONLY. Every other instruction in this prompt —
location, time of day, people and their identity, poses, props,
carried state, physical rules, and the no-text rule — is binding
exactly as written earlier. If the VISUAL APPROACH appears to conflict
with any of them, the earlier instruction wins.

Edit this photograph. Apply ONLY the corrections listed below,
each in place, changing nothing else:

CORRECTIONS:
- Remove the unrequested titles and misspelled Korean labels from the monitor screen, leaving only the specified coordinates, timestamp, and status indicators.

PRESERVE EVERYTHING ELSE EXACTLY as in the original photograph —
every person, pose, face, garment, object, set feature, lighting,
camera and framing stay identical. No text, captions or
watermarks anywhere.

Create ONE FINAL photorealistic live-action film still of the moment below — contemporary South Korea, 2026; all people are Korean unless stated. TIME OF DAY (lock): dusk.

SHOT TEXT (authoritative, Korean): 서랍이 열려 있고 식탁 의자가 바닥에 나뒹구는 난장판이 된 거실 전경

LOCATION (lock): Inside the ransacked rooftop apartment living and dining room at dusk, with the window open and furniture overturned. The shot takes place here — the attached LOCATION PHOTOGRAPH shows the exact spot.

CAMERA & FRAME (follow exactly — this is the composition authority for this still; the reference images supply identity and place, never the framing):
- CAMERA: From just inside the entrance at chest height, the camera briefly settles during its forward dolly into a wide view across the entire living room. Open drawers spread through the midground, fallen dining chairs break up the lower frame, and the open window remains visible beyond them, preserving a clear route toward the bedroom side of the room.
- FRAMING SCALE: wide shot
- KEY BACKGROUND ELEMENTS: Open drawers (Open and emptied or disturbed) — Their pulled-out fronts face the entrance at varied angles; used as Repeated midground shapes that guide the eye across the searched room; Dining chairs (Fallen on the floor) — Their seats, backs, and upturned legs face different directions across the floor; used as Lower-frame obstacles that make the room feel difficult to cross; Living-room window (Open) — The room-facing side and open aperture are visible from the entrance; used as A background opening that extends the room's depth; Footprints (Visible across the room); used as Visible traces connecting the entrance to the disturbed interior.
Compose the frame exactly as specified above — camera angle, subject scale and screen placement. Keep true physical scale between people and background elements: fixtures and distant objects occupy only the small screen area their real size and distance dictate; never enlarge a background object into a foreground presence.

REALIZE FIGURATIVE LANGUAGE AS A LIVE-ACTION SHOT: the Korean shot
text may describe characters metaphorically, figuratively or with
exaggeration. Photograph what a real movie camera would actually
record on a physical set — exaggerated or figurative impressions
become realistic staging choices (lighting, distance, angle,
wardrobe), not literal fantasy imagery.
EVERY CHARACTER IS A HUMAN BEING: unless the story explicitly
features non-human or virtual beings (as in science fiction or
fantasy), every character — however indirectly, vaguely or
figuratively the text describes them — IS a real human. When the
text gives no direct visual description of a person, IMAGINE one
and still show them as a concrete, fully-formed human being:
always render the human form to the maximum extent — build,
posture, face, hands, clothing — never reduce a person to a shape,
blob, solid silhouette or abstract mass.

EXPRESSIONS ARE ACTED, NEVER ANATOMICAL: when the text describes a
person's eyes, face or presence emotionally or figuratively (vacant,
hollow, dazed, burning, lifeless gaze and the like), it describes an
ACTOR'S PERFORMANCE captured by a real camera — realize it ONLY
through gaze direction, focus, eyelids, facial muscles, stillness
and posture. Every living person's eyes remain anatomically normal
human eyes with a natural iris and pupil, natural sclera and normal
proportions; NEVER whiten, blank out, cloud over, glow, enlarge or
otherwise alter eyeballs, skin or anatomy — unless the story
explicitly declares that being non-human or supernatural in form.

PROPS FACE THE RIGHT WAY: every handheld or used object must be
oriented exactly as its real-world use requires. A person reading,
watching or operating something (a phone, a photograph, a paper,
any device) has its functional side — screen, front, page — facing
THEIR OWN eyes; the camera then sees whatever side the staging
geometry implies (often its back). Show the functional side to the
camera ONLY when the shot text itself stages it toward the viewer.
Never flip, mirror or reverse an object's front and back.

CARRIED STATE (persist exactly — must match the neighbouring shots of this scene): The home has been ransacked, leaving footprints, open drawers, an open living-room window, and overturned dining chairs. The red circle and bloodied old photograph are already present on Su-ri-young’s bedroom wall, while her backpack still contains the printed maps.

NO PEOPLE IN THIS SHOT: the shot text shows only the place and its state — no living person or any body part appears in frame, unless the shot text itself explicitly says so.

No text, captions, watermarks or annotations anywhere.

Create ONE FINAL photorealistic live-action film still of the moment below — contemporary South Korea, 2026; all people are Korean unless stated. TIME OF DAY (lock): dusk.

SHOT TEXT (authoritative, Korean): 서랍이 열려 있고 식탁 의자가 바닥에 나뒹구는 난장판이 된 거실 전경

LOCATION (lock): Inside the ransacked rooftop apartment living and dining room at dusk, with the window open and furniture overturned. The shot takes place here — the attached LOCATION PHOTOGRAPH shows the exact spot.

CAMERA & FRAME (follow exactly — this is the composition authority for this still; the reference images supply identity and place, never the framing):
- CAMERA: From just inside the entrance at chest height, the camera briefly settles during its forward dolly into a wide view across the entire living room. Open drawers spread through the midground, fallen dining chairs break up the lower frame, and the open window remains visible beyond them, preserving a clear route toward the bedroom side of the room.
- FRAMING SCALE: wide shot
- KEY BACKGROUND ELEMENTS: Open drawers (Open and emptied or disturbed) — Their pulled-out fronts face the entrance at varied angles; used as Repeated midground shapes that guide the eye across the searched room; Dining chairs (Fallen on the floor) — Their seats, backs, and upturned legs face different directions across the floor; used as Lower-frame obstacles that make the room feel difficult to cross; Living-room window (Open) — The room-facing side and open aperture are visible from the entrance; used as A background opening that extends the room's depth; Footprints (Visible across the room); used as Visible traces connecting the entrance to the disturbed interior.
Compose the frame exactly as specified above — camera angle, subject scale and screen placement. Keep true physical scale between people and background elements: fixtures and distant objects occupy only the small screen area their real size and distance dictate; never enlarge a background object into a foreground presence.

REALIZE FIGURATIVE LANGUAGE AS A LIVE-ACTION SHOT: the Korean shot
text may describe characters metaphorically, figuratively or with
exaggeration. Photograph what a real movie camera would actually
record on a physical set — exaggerated or figurative impressions
become realistic staging choices (lighting, distance, angle,
wardrobe), not literal fantasy imagery.
EVERY CHARACTER IS A HUMAN BEING: unless the story explicitly
features non-human or virtual beings (as in science fiction or
fantasy), every character — however indirectly, vaguely or
figuratively the text describes them — IS a real human. When the
text gives no direct visual description of a person, IMAGINE one
and still show them as a concrete, fully-formed human being:
always render the human form to the maximum extent — build,
posture, face, hands, clothing — never reduce a person to a shape,
blob, solid silhouette or abstract mass.

EXPRESSIONS ARE ACTED, NEVER ANATOMICAL: when the text describes a
person's eyes, face or presence emotionally or figuratively (vacant,
hollow, dazed, burning, lifeless gaze and the like), it describes an
ACTOR'S PERFORMANCE captured by a real camera — realize it ONLY
through gaze direction, focus, eyelids, facial muscles, stillness
and posture. Every living person's eyes remain anatomically normal
human eyes with a natural iris and pupil, natural sclera and normal
proportions; NEVER whiten, blank out, cloud over, glow, enlarge or
otherwise alter eyeballs, skin or anatomy — unless the story
explicitly declares that being non-human or supernatural in form.

PROPS FACE THE RIGHT WAY: every handheld or used object must be
oriented exactly as its real-world use requires. A person reading,
watching or operating something (a phone, a photograph, a paper,
any device) has its functional side — screen, front, page — facing
THEIR OWN eyes; the camera then sees whatever side the staging
geometry implies (often its back). Show the functional side to the
camera ONLY when the shot text itself stages it toward the viewer.
Never flip, mirror or reverse an object's front and back.

CARRIED STATE (persist exactly — must match the neighbouring shots of this scene): The home has been ransacked, leaving footprints, open drawers, an open living-room window, and overturned dining chairs. The red circle and bloodied old photograph are already present on Su-ri-young’s bedroom wall, while her backpack still contains the printed maps.

NO PEOPLE IN THIS SHOT: the shot text shows only the place and its state — no living person or any body part appears in frame, unless the shot text itself explicitly says so.

No text, captions, watermarks or annotations anywhere.

VISUAL APPROACH — this take (camera, framing, composition):
Photograph the room with a restrained, nearly head-on wide perspective and deep focus, letting the repeated open drawers and scattered chair forms read with forensic clarity from foreground to background. Use the open window as the cool dusk-lit depth anchor, with soft ambient falloff through the interior; keep the footprints legible as a continuous trace through the layered disorder. The settled composition should feel cold, observational, and unnervingly still.

PRIORITY OF INSTRUCTIONS: the VISUAL APPROACH above governs camera,
framing and composition ONLY. Every other instruction in this prompt —
location, time of day, people and their identity, poses, props,
carried state, physical rules, and the no-text rule — is binding
exactly as written earlier. If the VISUAL APPROACH appears to conflict
with any of them, the earlier instruction wins.

Create ONE FINAL photorealistic live-action film still of the moment below — contemporary South Korea, 2026; all people are Korean unless stated. TIME OF DAY (lock): dusk.

SHOT TEXT (authoritative, Korean): 서랍이 열려 있고 식탁 의자가 바닥에 나뒹구는 난장판이 된 거실 전경

LOCATION (lock): Inside the ransacked rooftop apartment living and dining room at dusk, with the window open and furniture overturned. The shot takes place here — the attached LOCATION PHOTOGRAPH shows the exact spot.

CAMERA & FRAME (follow exactly — this is the composition authority for this still; the reference images supply identity and place, never the framing):
- CAMERA: From just inside the entrance at chest height, the camera briefly settles during its forward dolly into a wide view across the entire living room. Open drawers spread through the midground, fallen dining chairs break up the lower frame, and the open window remains visible beyond them, preserving a clear route toward the bedroom side of the room.
- FRAMING SCALE: wide shot
- KEY BACKGROUND ELEMENTS: Open drawers (Open and emptied or disturbed) — Their pulled-out fronts face the entrance at varied angles; used as Repeated midground shapes that guide the eye across the searched room; Dining chairs (Fallen on the floor) — Their seats, backs, and upturned legs face different directions across the floor; used as Lower-frame obstacles that make the room feel difficult to cross; Living-room window (Open) — The room-facing side and open aperture are visible from the entrance; used as A background opening that extends the room's depth; Footprints (Visible across the room); used as Visible traces connecting the entrance to the disturbed interior.
Compose the frame exactly as specified above — camera angle, subject scale and screen placement. Keep true physical scale between people and background elements: fixtures and distant objects occupy only the small screen area their real size and distance dictate; never enlarge a background object into a foreground presence.

REALIZE FIGURATIVE LANGUAGE AS A LIVE-ACTION SHOT: the Korean shot
text may describe characters metaphorically, figuratively or with
exaggeration. Photograph what a real movie camera would actually
record on a physical set — exaggerated or figurative impressions
become realistic staging choices (lighting, distance, angle,
wardrobe), not literal fantasy imagery.
EVERY CHARACTER IS A HUMAN BEING: unless the story explicitly
features non-human or virtual beings (as in science fiction or
fantasy), every character — however indirectly, vaguely or
figuratively the text describes them — IS a real human. When the
text gives no direct visual description of a person, IMAGINE one
and still show them as a concrete, fully-formed human being:
always render the human form to the maximum extent — build,
posture, face, hands, clothing — never reduce a person to a shape,
blob, solid silhouette or abstract mass.

EXPRESSIONS ARE ACTED, NEVER ANATOMICAL: when the text describes a
person's eyes, face or presence emotionally or figuratively (vacant,
hollow, dazed, burning, lifeless gaze and the like), it describes an
ACTOR'S PERFORMANCE captured by a real camera — realize it ONLY
through gaze direction, focus, eyelids, facial muscles, stillness
and posture. Every living person's eyes remain anatomically normal
human eyes with a natural iris and pupil, natural sclera and normal
proportions; NEVER whiten, blank out, cloud over, glow, enlarge or
otherwise alter eyeballs, skin or anatomy — unless the story
explicitly declares that being non-human or supernatural in form.

PROPS FACE THE RIGHT WAY: every handheld or used object must be
oriented exactly as its real-world use requires. A person reading,
watching or operating something (a phone, a photograph, a paper,
any device) has its functional side — screen, front, page — facing
THEIR OWN eyes; the camera then sees whatever side the staging
geometry implies (often its back). Show the functional side to the
camera ONLY when the shot text itself stages it toward the viewer.
Never flip, mirror or reverse an object's front and back.

CARRIED STATE (persist exactly — must match the neighbouring shots of this scene): The home has been ransacked, leaving footprints, open drawers, an open living-room window, and overturned dining chairs. The red circle and bloodied old photograph are already present on Su-ri-young’s bedroom wall, while her backpack still contains the printed maps.

NO PEOPLE IN THIS SHOT: the shot text shows only the place and its state — no living person or any body part appears in frame, unless the shot text itself explicitly says so.

No text, captions, watermarks or annotations anywhere.

VISUAL APPROACH — this take (camera, framing, composition):
Build an uneasy diagonal composition in which the fallen dining chairs press prominently into the lower frame while the disturbed drawers stagger across the midground toward the open window. Favor a slightly wider, more immersive lens feel with strong near-to-far perspective, preserving the clear route toward the bedroom side while making the room feel physically treacherous to cross. Let cool dusk illumination from the window skim unevenly across the floor so the footprints emerge in broken visual beats, with modest depth falloff that retains all essential background forms.

PRIORITY OF INSTRUCTIONS: the VISUAL APPROACH above governs camera,
framing and composition ONLY. Every other instruction in this prompt —
location, time of day, people and their identity, poses, props,
carried state, physical rules, and the no-text rule — is binding
exactly as written earlier. If the VISUAL APPROACH appears to conflict
with any of them, the earlier instruction wins.

Edit this photograph. Apply ONLY the corrections listed below,
each in place, changing nothing else:

CORRECTIONS:
- Overturn the upright chair at the dining table so it lies fallen on the floor.
- Change the outward-swinging window panes back to a sliding glass window matching the reference.

PRESERVE EVERYTHING ELSE EXACTLY as in the original photograph —
every person, pose, face, garment, object, set feature, lighting,
camera and framing stay identical. No text, captions or
watermarks anywhere.

Create ONE FINAL photorealistic live-action film still of the moment below — contemporary South Korea, 2026; all people are Korean unless stated. TIME OF DAY (lock): dusk.

SHOT TEXT (authoritative, Korean): 수리영 방 벽면에 그려진 거대한 붉은 원과 그 중앙에 꽂힌 낡은 사진 한 장의 모습

LOCATION (lock): Inside the daughter's bedroom at dusk, focused on the wall bearing a large red circle and a photograph pinned at its center. The shot takes place here — the attached LOCATION PHOTOGRAPH shows the exact spot.

CAMERA & FRAME (follow exactly — this is the composition authority for this still; the reference images supply identity and place, never the framing):
- CAMERA: At eye level inside the bedroom, the dolly-in completes within arm’s reach of the wall, framing the large red circle while keeping the photograph at its exact center. The photograph occupies roughly a quarter of the close frame rather than eclipsing the wall mark, creating a graphic visual bridge into the depicted memory.
- FRAMING SCALE: close-up
- FRAME LAYOUT: old photograph at the circle's center in the middle-center of the frame, midground; large red circle in the middle-center of the frame, background.
- KEY BACKGROUND ELEMENTS: Bedroom wall marked with a red circle (Marked with a large red circle) — The camera-facing wall plane bears the full large red circle, centered in frame; used as Graphic frame and crimson focal accent surrounding the photograph; Old photograph (Fixed at the circle's center and bloodied) — Its image-bearing front faces the camera, showing a rural restaurant, two young girls, and the circular structure on the roof; used as Central focal point and match-cut source for the following memory image.
Compose the frame exactly as specified above — camera angle, subject scale and screen placement. Keep true physical scale between people and background elements: fixtures and distant objects occupy only the small screen area their real size and distance dictate; never enlarge a background object into a foreground presence.

REALIZE FIGURATIVE LANGUAGE AS A LIVE-ACTION SHOT: the Korean shot
text may describe characters metaphorically, figuratively or with
exaggeration. Photograph what a real movie camera would actually
record on a physical set — exaggerated or figurative impressions
become realistic staging choices (lighting, distance, angle,
wardrobe), not literal fantasy imagery.
EVERY CHARACTER IS A HUMAN BEING: unless the story explicitly
features non-human or virtual beings (as in science fiction or
fantasy), every character — however indirectly, vaguely or
figuratively the text describes them — IS a real human. When the
text gives no direct visual description of a person, IMAGINE one
and still show them as a concrete, fully-formed human being:
always render the human form to the maximum extent — build,
posture, face, hands, clothing — never reduce a person to a shape,
blob, solid silhouette or abstract mass.

EXPRESSIONS ARE ACTED, NEVER ANATOMICAL: when the text describes a
person's eyes, face or presence emotionally or figuratively (vacant,
hollow, dazed, burning, lifeless gaze and the like), it describes an
ACTOR'S PERFORMANCE captured by a real camera — realize it ONLY
through gaze direction, focus, eyelids, facial muscles, stillness
and posture. Every living person's eyes remain anatomically normal
human eyes with a natural iris and pupil, natural sclera and normal
proportions; NEVER whiten, blank out, cloud over, glow, enlarge or
otherwise alter eyeballs, skin or anatomy — unless the story
explicitly declares that being non-human or supernatural in form.

PROPS FACE THE RIGHT WAY: every handheld or used object must be
oriented exactly as its real-world use requires. A person reading,
watching or operating something (a phone, a photograph, a paper,
any device) has its functional side — screen, front, page — facing
THEIR OWN eyes; the camera then sees whatever side the staging
geometry implies (often its back). Show the functional side to the
camera ONLY when the shot text itself stages it toward the viewer.
Never flip, mirror or reverse an object's front and back.

CARRIED STATE (persist exactly — must match the neighbouring shots of this scene): The ransacked state of the home persists, and a bloodied old photograph remains pinned at the center of the large red circle on the bedroom wall. Su-ri-young still has her map-filled backpack, coordinate note, and phone.

NO PEOPLE IN THIS SHOT: the shot text shows only the place and its state — no living person or any body part appears in frame, unless the shot text itself explicitly says so.

No text, captions, watermarks or annotations anywhere.

Create ONE FINAL photorealistic live-action film still of the moment below — contemporary South Korea, 2026; all people are Korean unless stated. TIME OF DAY (lock): dusk.

SHOT TEXT (authoritative, Korean): 수리영 방 벽면에 그려진 거대한 붉은 원과 그 중앙에 꽂힌 낡은 사진 한 장의 모습

LOCATION (lock): Inside the daughter's bedroom at dusk, focused on the wall bearing a large red circle and a photograph pinned at its center. The shot takes place here — the attached LOCATION PHOTOGRAPH shows the exact spot.

CAMERA & FRAME (follow exactly — this is the composition authority for this still; the reference images supply identity and place, never the framing):
- CAMERA: At eye level inside the bedroom, the dolly-in completes within arm’s reach of the wall, framing the large red circle while keeping the photograph at its exact center. The photograph occupies roughly a quarter of the close frame rather than eclipsing the wall mark, creating a graphic visual bridge into the depicted memory.
- FRAMING SCALE: close-up
- FRAME LAYOUT: old photograph at the circle's center in the middle-center of the frame, midground; large red circle in the middle-center of the frame, background.
- KEY BACKGROUND ELEMENTS: Bedroom wall marked with a red circle (Marked with a large red circle) — The camera-facing wall plane bears the full large red circle, centered in frame; used as Graphic frame and crimson focal accent surrounding the photograph; Old photograph (Fixed at the circle's center and bloodied) — Its image-bearing front faces the camera, showing a rural restaurant, two young girls, and the circular structure on the roof; used as Central focal point and match-cut source for the following memory image.
Compose the frame exactly as specified above — camera angle, subject scale and screen placement. Keep true physical scale between people and background elements: fixtures and distant objects occupy only the small screen area their real size and distance dictate; never enlarge a background object into a foreground presence.

REALIZE FIGURATIVE LANGUAGE AS A LIVE-ACTION SHOT: the Korean shot
text may describe characters metaphorically, figuratively or with
exaggeration. Photograph what a real movie camera would actually
record on a physical set — exaggerated or figurative impressions
become realistic staging choices (lighting, distance, angle,
wardrobe), not literal fantasy imagery.
EVERY CHARACTER IS A HUMAN BEING: unless the story explicitly
features non-human or virtual beings (as in science fiction or
fantasy), every character — however indirectly, vaguely or
figuratively the text describes them — IS a real human. When the
text gives no direct visual description of a person, IMAGINE one
and still show them as a concrete, fully-formed human being:
always render the human form to the maximum extent — build,
posture, face, hands, clothing — never reduce a person to a shape,
blob, solid silhouette or abstract mass.

EXPRESSIONS ARE ACTED, NEVER ANATOMICAL: when the text describes a
person's eyes, face or presence emotionally or figuratively (vacant,
hollow, dazed, burning, lifeless gaze and the like), it describes an
ACTOR'S PERFORMANCE captured by a real camera — realize it ONLY
through gaze direction, focus, eyelids, facial muscles, stillness
and posture. Every living person's eyes remain anatomically normal
human eyes with a natural iris and pupil, natural sclera and normal
proportions; NEVER whiten, blank out, cloud over, glow, enlarge or
otherwise alter eyeballs, skin or anatomy — unless the story
explicitly declares that being non-human or supernatural in form.

PROPS FACE THE RIGHT WAY: every handheld or used object must be
oriented exactly as its real-world use requires. A person reading,
watching or operating something (a phone, a photograph, a paper,
any device) has its functional side — screen, front, page — facing
THEIR OWN eyes; the camera then sees whatever side the staging
geometry implies (often its back). Show the functional side to the
camera ONLY when the shot text itself stages it toward the viewer.
Never flip, mirror or reverse an object's front and back.

CARRIED STATE (persist exactly — must match the neighbouring shots of this scene): The ransacked state of the home persists, and a bloodied old photograph remains pinned at the center of the large red circle on the bedroom wall. Su-ri-young still has her map-filled backpack, coordinate note, and phone.

NO PEOPLE IN THIS SHOT: the shot text shows only the place and its state — no living person or any body part appears in frame, unless the shot text itself explicitly says so.

No text, captions, watermarks or annotations anywhere.

VISUAL APPROACH — this take (camera, framing, composition):
Photograph the wall square-on with rigid, near-perfect symmetry and a restrained normal-lens perspective. Keep the photograph and the surrounding circle crisply legible across the shallow spatial plane, using soft, cool dusk ambience and controlled contrast so the crimson mark dominates as a stark graphic field while the aged, bloodied surface of the photograph remains tactile and precise.

PRIORITY OF INSTRUCTIONS: the VISUAL APPROACH above governs camera,
framing and composition ONLY. Every other instruction in this prompt —
location, time of day, people and their identity, poses, props,
carried state, physical rules, and the no-text rule — is binding
exactly as written earlier. If the VISUAL APPROACH appears to conflict
with any of them, the earlier instruction wins.

Create ONE FINAL photorealistic live-action film still of the moment below — contemporary South Korea, 2026; all people are Korean unless stated. TIME OF DAY (lock): dusk.

SHOT TEXT (authoritative, Korean): 수리영 방 벽면에 그려진 거대한 붉은 원과 그 중앙에 꽂힌 낡은 사진 한 장의 모습

LOCATION (lock): Inside the daughter's bedroom at dusk, focused on the wall bearing a large red circle and a photograph pinned at its center. The shot takes place here — the attached LOCATION PHOTOGRAPH shows the exact spot.

CAMERA & FRAME (follow exactly — this is the composition authority for this still; the reference images supply identity and place, never the framing):
- CAMERA: At eye level inside the bedroom, the dolly-in completes within arm’s reach of the wall, framing the large red circle while keeping the photograph at its exact center. The photograph occupies roughly a quarter of the close frame rather than eclipsing the wall mark, creating a graphic visual bridge into the depicted memory.
- FRAMING SCALE: close-up
- FRAME LAYOUT: old photograph at the circle's center in the middle-center of the frame, midground; large red circle in the middle-center of the frame, background.
- KEY BACKGROUND ELEMENTS: Bedroom wall marked with a red circle (Marked with a large red circle) — The camera-facing wall plane bears the full large red circle, centered in frame; used as Graphic frame and crimson focal accent surrounding the photograph; Old photograph (Fixed at the circle's center and bloodied) — Its image-bearing front faces the camera, showing a rural restaurant, two young girls, and the circular structure on the roof; used as Central focal point and match-cut source for the following memory image.
Compose the frame exactly as specified above — camera angle, subject scale and screen placement. Keep true physical scale between people and background elements: fixtures and distant objects occupy only the small screen area their real size and distance dictate; never enlarge a background object into a foreground presence.

REALIZE FIGURATIVE LANGUAGE AS A LIVE-ACTION SHOT: the Korean shot
text may describe characters metaphorically, figuratively or with
exaggeration. Photograph what a real movie camera would actually
record on a physical set — exaggerated or figurative impressions
become realistic staging choices (lighting, distance, angle,
wardrobe), not literal fantasy imagery.
EVERY CHARACTER IS A HUMAN BEING: unless the story explicitly
features non-human or virtual beings (as in science fiction or
fantasy), every character — however indirectly, vaguely or
figuratively the text describes them — IS a real human. When the
text gives no direct visual description of a person, IMAGINE one
and still show them as a concrete, fully-formed human being:
always render the human form to the maximum extent — build,
posture, face, hands, clothing — never reduce a person to a shape,
blob, solid silhouette or abstract mass.

EXPRESSIONS ARE ACTED, NEVER ANATOMICAL: when the text describes a
person's eyes, face or presence emotionally or figuratively (vacant,
hollow, dazed, burning, lifeless gaze and the like), it describes an
ACTOR'S PERFORMANCE captured by a real camera — realize it ONLY
through gaze direction, focus, eyelids, facial muscles, stillness
and posture. Every living person's eyes remain anatomically normal
human eyes with a natural iris and pupil, natural sclera and normal
proportions; NEVER whiten, blank out, cloud over, glow, enlarge or
otherwise alter eyeballs, skin or anatomy — unless the story
explicitly declares that being non-human or supernatural in form.

PROPS FACE THE RIGHT WAY: every handheld or used object must be
oriented exactly as its real-world use requires. A person reading,
watching or operating something (a phone, a photograph, a paper,
any device) has its functional side — screen, front, page — facing
THEIR OWN eyes; the camera then sees whatever side the staging
geometry implies (often its back). Show the functional side to the
camera ONLY when the shot text itself stages it toward the viewer.
Never flip, mirror or reverse an object's front and back.

CARRIED STATE (persist exactly — must match the neighbouring shots of this scene): The ransacked state of the home persists, and a bloodied old photograph remains pinned at the center of the large red circle on the bedroom wall. Su-ri-young still has her map-filled backpack, coordinate note, and phone.

NO PEOPLE IN THIS SHOT: the shot text shows only the place and its state — no living person or any body part appears in frame, unless the shot text itself explicitly says so.

No text, captions, watermarks or annotations anywhere.

VISUAL APPROACH — this take (camera, framing, composition):
Use an intimate, slightly wider close-focus lens feel, with the photograph rendered in piercing sharpness and the complete red circle falling into a subtly softer wall plane around it. Let dim dusk light rake gently across the surface, revealing pinning, stains, paper wear, and wall texture; allow restrained edge falloff to draw the eye inward while preserving the exact centered composition and the photograph’s match-cut clarity.

PRIORITY OF INSTRUCTIONS: the VISUAL APPROACH above governs camera,
framing and composition ONLY. Every other instruction in this prompt —
location, time of day, people and their identity, poses, props,
carried state, physical rules, and the no-text rule — is binding
exactly as written earlier. If the VISUAL APPROACH appears to conflict
with any of them, the earlier instruction wins.

Edit this photograph. Apply ONLY the corrections listed below,
each in place, changing nothing else:

CORRECTIONS:
- Remove all text from the sign on the restaurant building in the photograph.

PRESERVE EVERYTHING ELSE EXACTLY as in the original photograph —
every person, pose, face, garment, object, set feature, lighting,
camera and framing stay identical. No text, captions or
watermarks anywhere.

Create ONE FINAL photorealistic live-action film still of the moment below — contemporary South Korea, 2026; all people are Korean unless stated. TIME OF DAY (lock): sunset, foggy.

SHOT TEXT (authoritative, Korean): 소형 어선의 뱃머리가 짙은 안개벽 표면에 맞닿은 찰나의 전경

LOCATION (lock): On the distant open sea at sunset, where the bow of the small fishing boat meets a dense wall of fog. The shot takes place here — the attached LOCATION PHOTOGRAPH shows the exact spot.

CAMERA & FRAME (follow exactly — this is the composition authority for this still; the reference images supply identity and place, never the framing):
- CAMERA: Low above the sea and slightly ahead of the vessel, the camera tracks backward while facing across the bow at the exact instant its tip meets the dense fog boundary. The bow occupies no more than the lower third as a scale anchor, while the fog band spans the upper field and the last visible strip of sea separates and then closes between them.
- FRAMING SCALE: medium shot
- FRAME LAYOUT: fishing boat bow in the lower-center of the frame, foreground, moves toward dense fog boundary; dense fog boundary in the upper-center of the frame, background.
- KEY BACKGROUND ELEMENTS: small fishing boat bow (Its leading edge is touching the fog boundary) — The front and one oblique side face the camera from the low forward-side position; used as Foreground scale anchor whose forward edge marks the instant of contact; dense fog bank (Dense and erasing the view beyond it); used as Primary background boundary into which the vessel continues; sea surface (Visible below and beside the bow) — The surface reflects the golden cast of the setting sun around the boat's approach line; used as Lower-frame spatial gap and reflective field beneath the approaching bow; setting sun (Gradually descending); used as Source of the scene's explicitly described golden cast across the sea.
Compose the frame exactly as specified above — camera angle, subject scale and screen placement. Keep true physical scale between people and background elements: fixtures and distant objects occupy only the small screen area their real size and distance dictate; never enlarge a background object into a foreground presence.

REALIZE FIGURATIVE LANGUAGE AS A LIVE-ACTION SHOT: the Korean shot
text may describe characters metaphorically, figuratively or with
exaggeration. Photograph what a real movie camera would actually
record on a physical set — exaggerated or figurative impressions
become realistic staging choices (lighting, distance, angle,
wardrobe), not literal fantasy imagery.
EVERY CHARACTER IS A HUMAN BEING: unless the story explicitly
features non-human or virtual beings (as in science fiction or
fantasy), every character — however indirectly, vaguely or
figuratively the text describes them — IS a real human. When the
text gives no direct visual description of a person, IMAGINE one
and still show them as a concrete, fully-formed human being:
always render the human form to the maximum extent — build,
posture, face, hands, clothing — never reduce a person to a shape,
blob, solid silhouette or abstract mass.

EXPRESSIONS ARE ACTED, NEVER ANATOMICAL: when the text describes a
person's eyes, face or presence emotionally or figuratively (vacant,
hollow, dazed, burning, lifeless gaze and the like), it describes an
ACTOR'S PERFORMANCE captured by a real camera — realize it ONLY
through gaze direction, focus, eyelids, facial muscles, stillness
and posture. Every living person's eyes remain anatomically normal
human eyes with a natural iris and pupil, natural sclera and normal
proportions; NEVER whiten, blank out, cloud over, glow, enlarge or
otherwise alter eyeballs, skin or anatomy — unless the story
explicitly declares that being non-human or supernatural in form.

PROPS FACE THE RIGHT WAY: every handheld or used object must be
oriented exactly as its real-world use requires. A person reading,
watching or operating something (a phone, a photograph, a paper,
any device) has its functional side — screen, front, page — facing
THEIR OWN eyes; the camera then sees whatever side the staging
geometry implies (often its back). Show the functional side to the
camera ONLY when the shot text itself stages it toward the viewer.
Never flip, mirror or reverse an object's front and back.

CARRIED STATE (persist exactly — must match the neighbouring shots of this scene): The same small fishing boat enters the dense fog with Su-ri-young still unconscious and slumped on In-woo inside the wheelhouse. Her photograph, backpack, maps, coordinate note, and phone remain aboard.

NO PEOPLE IN THIS SHOT: the shot text shows only the place and its state — no living person or any body part appears in frame, unless the shot text itself explicitly says so.

No text, captions, watermarks or annotations anywhere.

Create ONE FINAL photorealistic live-action film still of the moment below — contemporary South Korea, 2026; all people are Korean unless stated. TIME OF DAY (lock): sunset, foggy.

SHOT TEXT (authoritative, Korean): 소형 어선의 뱃머리가 짙은 안개벽 표면에 맞닿은 찰나의 전경

LOCATION (lock): On the distant open sea at sunset, where the bow of the small fishing boat meets a dense wall of fog. The shot takes place here — the attached LOCATION PHOTOGRAPH shows the exact spot.

CAMERA & FRAME (follow exactly — this is the composition authority for this still; the reference images supply identity and place, never the framing):
- CAMERA: Low above the sea and slightly ahead of the vessel, the camera tracks backward while facing across the bow at the exact instant its tip meets the dense fog boundary. The bow occupies no more than the lower third as a scale anchor, while the fog band spans the upper field and the last visible strip of sea separates and then closes between them.
- FRAMING SCALE: medium shot
- FRAME LAYOUT: fishing boat bow in the lower-center of the frame, foreground, moves toward dense fog boundary; dense fog boundary in the upper-center of the frame, background.
- KEY BACKGROUND ELEMENTS: small fishing boat bow (Its leading edge is touching the fog boundary) — The front and one oblique side face the camera from the low forward-side position; used as Foreground scale anchor whose forward edge marks the instant of contact; dense fog bank (Dense and erasing the view beyond it); used as Primary background boundary into which the vessel continues; sea surface (Visible below and beside the bow) — The surface reflects the golden cast of the setting sun around the boat's approach line; used as Lower-frame spatial gap and reflective field beneath the approaching bow; setting sun (Gradually descending); used as Source of the scene's explicitly described golden cast across the sea.
Compose the frame exactly as specified above — camera angle, subject scale and screen placement. Keep true physical scale between people and background elements: fixtures and distant objects occupy only the small screen area their real size and distance dictate; never enlarge a background object into a foreground presence.

REALIZE FIGURATIVE LANGUAGE AS A LIVE-ACTION SHOT: the Korean shot
text may describe characters metaphorically, figuratively or with
exaggeration. Photograph what a real movie camera would actually
record on a physical set — exaggerated or figurative impressions
become realistic staging choices (lighting, distance, angle,
wardrobe), not literal fantasy imagery.
EVERY CHARACTER IS A HUMAN BEING: unless the story explicitly
features non-human or virtual beings (as in science fiction or
fantasy), every character — however indirectly, vaguely or
figuratively the text describes them — IS a real human. When the
text gives no direct visual description of a person, IMAGINE one
and still show them as a concrete, fully-formed human being:
always render the human form to the maximum extent — build,
posture, face, hands, clothing — never reduce a person to a shape,
blob, solid silhouette or abstract mass.

EXPRESSIONS ARE ACTED, NEVER ANATOMICAL: when the text describes a
person's eyes, face or presence emotionally or figuratively (vacant,
hollow, dazed, burning, lifeless gaze and the like), it describes an
ACTOR'S PERFORMANCE captured by a real camera — realize it ONLY
through gaze direction, focus, eyelids, facial muscles, stillness
and posture. Every living person's eyes remain anatomically normal
human eyes with a natural iris and pupil, natural sclera and normal
proportions; NEVER whiten, blank out, cloud over, glow, enlarge or
otherwise alter eyeballs, skin or anatomy — unless the story
explicitly declares that being non-human or supernatural in form.

PROPS FACE THE RIGHT WAY: every handheld or used object must be
oriented exactly as its real-world use requires. A person reading,
watching or operating something (a phone, a photograph, a paper,
any device) has its functional side — screen, front, page — facing
THEIR OWN eyes; the camera then sees whatever side the staging
geometry implies (often its back). Show the functional side to the
camera ONLY when the shot text itself stages it toward the viewer.
Never flip, mirror or reverse an object's front and back.

CARRIED STATE (persist exactly — must match the neighbouring shots of this scene): The same small fishing boat enters the dense fog with Su-ri-young still unconscious and slumped on In-woo inside the wheelhouse. Her photograph, backpack, maps, coordinate note, and phone remain aboard.

NO PEOPLE IN THIS SHOT: the shot text shows only the place and its state — no living person or any body part appears in frame, unless the shot text itself explicitly says so.

No text, captions, watermarks or annotations anywhere.

VISUAL APPROACH — this take (camera, framing, composition):
Use a moderately wide, naturalistic lens feel from the prescribed low forward-side position, letting the bow’s oblique lines converge forcefully toward the point of contact. Keep the sea-to-fog depth plainly legible, with crisp foreground texture transitioning into softly diffused density; low sunset light skims the reflective water while the fog returns a broad, muted golden glow. The image should feel immediate, physical, and quietly ominous.

PRIORITY OF INSTRUCTIONS: the VISUAL APPROACH above governs camera,
framing and composition ONLY. Every other instruction in this prompt —
location, time of day, people and their identity, poses, props,
carried state, physical rules, and the no-text rule — is binding
exactly as written earlier. If the VISUAL APPROACH appears to conflict
with any of them, the earlier instruction wins.

Create ONE FINAL photorealistic live-action film still of the moment below — contemporary South Korea, 2026; all people are Korean unless stated. TIME OF DAY (lock): sunset, foggy.

SHOT TEXT (authoritative, Korean): 소형 어선의 뱃머리가 짙은 안개벽 표면에 맞닿은 찰나의 전경

LOCATION (lock): On the distant open sea at sunset, where the bow of the small fishing boat meets a dense wall of fog. The shot takes place here — the attached LOCATION PHOTOGRAPH shows the exact spot.

CAMERA & FRAME (follow exactly — this is the composition authority for this still; the reference images supply identity and place, never the framing):
- CAMERA: Low above the sea and slightly ahead of the vessel, the camera tracks backward while facing across the bow at the exact instant its tip meets the dense fog boundary. The bow occupies no more than the lower third as a scale anchor, while the fog band spans the upper field and the last visible strip of sea separates and then closes between them.
- FRAMING SCALE: medium shot
- FRAME LAYOUT: fishing boat bow in the lower-center of the frame, foreground, moves toward dense fog boundary; dense fog boundary in the upper-center of the frame, background.
- KEY BACKGROUND ELEMENTS: small fishing boat bow (Its leading edge is touching the fog boundary) — The front and one oblique side face the camera from the low forward-side position; used as Foreground scale anchor whose forward edge marks the instant of contact; dense fog bank (Dense and erasing the view beyond it); used as Primary background boundary into which the vessel continues; sea surface (Visible below and beside the bow) — The surface reflects the golden cast of the setting sun around the boat's approach line; used as Lower-frame spatial gap and reflective field beneath the approaching bow; setting sun (Gradually descending); used as Source of the scene's explicitly described golden cast across the sea.
Compose the frame exactly as specified above — camera angle, subject scale and screen placement. Keep true physical scale between people and background elements: fixtures and distant objects occupy only the small screen area their real size and distance dictate; never enlarge a background object into a foreground presence.

REALIZE FIGURATIVE LANGUAGE AS A LIVE-ACTION SHOT: the Korean shot
text may describe characters metaphorically, figuratively or with
exaggeration. Photograph what a real movie camera would actually
record on a physical set — exaggerated or figurative impressions
become realistic staging choices (lighting, distance, angle,
wardrobe), not literal fantasy imagery.
EVERY CHARACTER IS A HUMAN BEING: unless the story explicitly
features non-human or virtual beings (as in science fiction or
fantasy), every character — however indirectly, vaguely or
figuratively the text describes them — IS a real human. When the
text gives no direct visual description of a person, IMAGINE one
and still show them as a concrete, fully-formed human being:
always render the human form to the maximum extent — build,
posture, face, hands, clothing — never reduce a person to a shape,
blob, solid silhouette or abstract mass.

EXPRESSIONS ARE ACTED, NEVER ANATOMICAL: when the text describes a
person's eyes, face or presence emotionally or figuratively (vacant,
hollow, dazed, burning, lifeless gaze and the like), it describes an
ACTOR'S PERFORMANCE captured by a real camera — realize it ONLY
through gaze direction, focus, eyelids, facial muscles, stillness
and posture. Every living person's eyes remain anatomically normal
human eyes with a natural iris and pupil, natural sclera and normal
proportions; NEVER whiten, blank out, cloud over, glow, enlarge or
otherwise alter eyeballs, skin or anatomy — unless the story
explicitly declares that being non-human or supernatural in form.

PROPS FACE THE RIGHT WAY: every handheld or used object must be
oriented exactly as its real-world use requires. A person reading,
watching or operating something (a phone, a photograph, a paper,
any device) has its functional side — screen, front, page — facing
THEIR OWN eyes; the camera then sees whatever side the staging
geometry implies (often its back). Show the functional side to the
camera ONLY when the shot text itself stages it toward the viewer.
Never flip, mirror or reverse an object's front and back.

CARRIED STATE (persist exactly — must match the neighbouring shots of this scene): The same small fishing boat enters the dense fog with Su-ri-young still unconscious and slumped on In-woo inside the wheelhouse. Her photograph, backpack, maps, coordinate note, and phone remain aboard.

NO PEOPLE IN THIS SHOT: the shot text shows only the place and its state — no living person or any body part appears in frame, unless the shot text itself explicitly says so.

No text, captions, watermarks or annotations anywhere.

VISUAL APPROACH — this take (camera, framing, composition):
Use a restrained, compressed lens feel that visually presses the bow, the narrowing strip of sea, and the fog boundary into layered planes while preserving the specified scale and placement. Compose around the stark meeting of the sharp leading edge and the feature-erasing fog, with subdued boat detail against a luminous but opaque sunset haze. Favor calm tonal gradation and minimal spatial release, creating a suspended, irreversible threshold.

PRIORITY OF INSTRUCTIONS: the VISUAL APPROACH above governs camera,
framing and composition ONLY. Every other instruction in this prompt —
location, time of day, people and their identity, poses, props,
carried state, physical rules, and the no-text rule — is binding
exactly as written earlier. If the VISUAL APPROACH appears to conflict
with any of them, the earlier instruction wins.

Edit this photograph. Apply ONLY the corrections listed below,
each in place, changing nothing else:

CORRECTIONS:
- Move the boat forward across the water until its leading tip directly touches the dense fog boundary.
- Scale down the boat bow so it occupies no more than the lower third of the frame.

PRESERVE EVERYTHING ELSE EXACTLY as in the original photograph —
every person, pose, face, garment, object, set feature, lighting,
camera and framing stay identical. No text, captions or
watermarks anywhere.

Create ONE FINAL photorealistic live-action film still of the moment below — contemporary South Korea, 2026; all people are Korean unless stated. TIME OF DAY (lock): night.

SHOT TEXT (authoritative, Korean): 옥상에 도착한 수리영이 반쯤 열린 현관 철문을 멍하니 응시하며 굳어버린 전신

LOCATION (lock): On the villa rooftop at the top of the exterior stairs, directly before the rooftop apartment's half-open metal entrance door. The shot takes place here — the attached STORYBOARD SKETCH fixes the staging, camera and figure placement of this exact place. No location photograph is attached — build the location itself strictly from the location text above and the shot text, inventing nothing beyond them.

REALIZE FIGURATIVE LANGUAGE AS A LIVE-ACTION SHOT: the Korean shot
text may describe characters metaphorically, figuratively or with
exaggeration. Photograph what a real movie camera would actually
record on a physical set — exaggerated or figurative impressions
become realistic staging choices (lighting, distance, angle,
wardrobe), not literal fantasy imagery.
EVERY CHARACTER IS A HUMAN BEING: unless the story explicitly
features non-human or virtual beings (as in science fiction or
fantasy), every character — however indirectly, vaguely or
figuratively the text describes them — IS a real human. When the
text gives no direct visual description of a person, IMAGINE one
and still show them as a concrete, fully-formed human being:
always render the human form to the maximum extent — build,
posture, face, hands, clothing — never reduce a person to a shape,
blob, solid silhouette or abstract mass.

EXPRESSIONS ARE ACTED, NEVER ANATOMICAL: when the text describes a
person's eyes, face or presence emotionally or figuratively (vacant,
hollow, dazed, burning, lifeless gaze and the like), it describes an
ACTOR'S PERFORMANCE captured by a real camera — realize it ONLY
through gaze direction, focus, eyelids, facial muscles, stillness
and posture. Every living person's eyes remain anatomically normal
human eyes with a natural iris and pupil, natural sclera and normal
proportions; NEVER whiten, blank out, cloud over, glow, enlarge or
otherwise alter eyeballs, skin or anatomy — unless the story
explicitly declares that being non-human or supernatural in form.

PROPS FACE THE RIGHT WAY: every handheld or used object must be
oriented exactly as its real-world use requires. A person reading,
watching or operating something (a phone, a photograph, a paper,
any device) has its functional side — screen, front, page — facing
THEIR OWN eyes; the camera then sees whatever side the staging
geometry implies (often its back). Show the functional side to the
camera ONLY when the shot text itself stages it toward the viewer.
Never flip, mirror or reverse an object's front and back.

CARRIED STATE (persist exactly — must match the neighbouring shots of this scene): Su-ri-young’s bicycle remains hastily parked in the rear yard below, while the rooftop iron entrance door stands half open.

PEOPLE: the SHOT TEXT alone decides whether any person is visible in this shot. IF a person appears, they must be one of: 수리영 (한국인 20대 초반 여성, 긴 검은 생머리, 젊은 타원형 얼굴) — never anyone else, and never add a person the shot text does not show. IF only part of a person is in frame (a hand, arm, foot, back, silhouette), that body part belongs to the specific person the shot text names — its sex, age, build, skin and grooming must unmistakably match that person's profile above.

No text, captions, watermarks or annotations anywhere.

Create ONE FINAL photorealistic live-action film still of the moment below — contemporary South Korea, 2026; all people are Korean unless stated. TIME OF DAY (lock): night.

SHOT TEXT (authoritative, Korean): 옥상에 도착한 수리영이 반쯤 열린 현관 철문을 멍하니 응시하며 굳어버린 전신

LOCATION (lock): On the villa rooftop at the top of the exterior stairs, directly before the rooftop apartment's half-open metal entrance door. The shot takes place here — the attached STORYBOARD SKETCH fixes the staging, camera and figure placement of this exact place. No location photograph is attached — build the location itself strictly from the location text above and the shot text, inventing nothing beyond them.

REALIZE FIGURATIVE LANGUAGE AS A LIVE-ACTION SHOT: the Korean shot
text may describe characters metaphorically, figuratively or with
exaggeration. Photograph what a real movie camera would actually
record on a physical set — exaggerated or figurative impressions
become realistic staging choices (lighting, distance, angle,
wardrobe), not literal fantasy imagery.
EVERY CHARACTER IS A HUMAN BEING: unless the story explicitly
features non-human or virtual beings (as in science fiction or
fantasy), every character — however indirectly, vaguely or
figuratively the text describes them — IS a real human. When the
text gives no direct visual description of a person, IMAGINE one
and still show them as a concrete, fully-formed human being:
always render the human form to the maximum extent — build,
posture, face, hands, clothing — never reduce a person to a shape,
blob, solid silhouette or abstract mass.

EXPRESSIONS ARE ACTED, NEVER ANATOMICAL: when the text describes a
person's eyes, face or presence emotionally or figuratively (vacant,
hollow, dazed, burning, lifeless gaze and the like), it describes an
ACTOR'S PERFORMANCE captured by a real camera — realize it ONLY
through gaze direction, focus, eyelids, facial muscles, stillness
and posture. Every living person's eyes remain anatomically normal
human eyes with a natural iris and pupil, natural sclera and normal
proportions; NEVER whiten, blank out, cloud over, glow, enlarge or
otherwise alter eyeballs, skin or anatomy — unless the story
explicitly declares that being non-human or supernatural in form.

PROPS FACE THE RIGHT WAY: every handheld or used object must be
oriented exactly as its real-world use requires. A person reading,
watching or operating something (a phone, a photograph, a paper,
any device) has its functional side — screen, front, page — facing
THEIR OWN eyes; the camera then sees whatever side the staging
geometry implies (often its back). Show the functional side to the
camera ONLY when the shot text itself stages it toward the viewer.
Never flip, mirror or reverse an object's front and back.

CARRIED STATE (persist exactly — must match the neighbouring shots of this scene): Su-ri-young’s bicycle remains hastily parked in the rear yard below, while the rooftop iron entrance door stands half open.

PEOPLE: the SHOT TEXT alone decides whether any person is visible in this shot. IF a person appears, they must be one of: 수리영 (한국인 20대 초반 여성, 긴 검은 생머리, 젊은 타원형 얼굴) — never anyone else, and never add a person the shot text does not show. IF only part of a person is in frame (a hand, arm, foot, back, silhouette), that body part belongs to the specific person the shot text names — its sex, age, build, skin and grooming must unmistakably match that person's profile above.

No text, captions, watermarks or annotations anywhere.

VISUAL APPROACH — this take (camera, framing, composition):
Frame a wide, low rear three-quarter view from the top of the exterior stairs, holding Su-ri-young’s entire rigid body and the half-open metal entrance door in the same deep composition. Let the physical gap between her and the doorway dominate the frame, with restrained night illumination defining her posture, long hair, and the hard planes of the metal door. Keep the image still, tense, and observational.

PRIORITY OF INSTRUCTIONS: the VISUAL APPROACH above governs camera,
framing and composition ONLY. Every other instruction in this prompt —
location, time of day, people and their identity, poses, props,
carried state, physical rules, and the no-text rule — is binding
exactly as written earlier. If the VISUAL APPROACH appears to conflict
with any of them, the earlier instruction wins.

Create ONE FINAL photorealistic live-action film still of the moment below — contemporary South Korea, 2026; all people are Korean unless stated. TIME OF DAY (lock): night.

SHOT TEXT (authoritative, Korean): 옥상에 도착한 수리영이 반쯤 열린 현관 철문을 멍하니 응시하며 굳어버린 전신

LOCATION (lock): On the villa rooftop at the top of the exterior stairs, directly before the rooftop apartment's half-open metal entrance door. The shot takes place here — the attached STORYBOARD SKETCH fixes the staging, camera and figure placement of this exact place. No location photograph is attached — build the location itself strictly from the location text above and the shot text, inventing nothing beyond them.

REALIZE FIGURATIVE LANGUAGE AS A LIVE-ACTION SHOT: the Korean shot
text may describe characters metaphorically, figuratively or with
exaggeration. Photograph what a real movie camera would actually
record on a physical set — exaggerated or figurative impressions
become realistic staging choices (lighting, distance, angle,
wardrobe), not literal fantasy imagery.
EVERY CHARACTER IS A HUMAN BEING: unless the story explicitly
features non-human or virtual beings (as in science fiction or
fantasy), every character — however indirectly, vaguely or
figuratively the text describes them — IS a real human. When the
text gives no direct visual description of a person, IMAGINE one
and still show them as a concrete, fully-formed human being:
always render the human form to the maximum extent — build,
posture, face, hands, clothing — never reduce a person to a shape,
blob, solid silhouette or abstract mass.

EXPRESSIONS ARE ACTED, NEVER ANATOMICAL: when the text describes a
person's eyes, face or presence emotionally or figuratively (vacant,
hollow, dazed, burning, lifeless gaze and the like), it describes an
ACTOR'S PERFORMANCE captured by a real camera — realize it ONLY
through gaze direction, focus, eyelids, facial muscles, stillness
and posture. Every living person's eyes remain anatomically normal
human eyes with a natural iris and pupil, natural sclera and normal
proportions; NEVER whiten, blank out, cloud over, glow, enlarge or
otherwise alter eyeballs, skin or anatomy — unless the story
explicitly declares that being non-human or supernatural in form.

PROPS FACE THE RIGHT WAY: every handheld or used object must be
oriented exactly as its real-world use requires. A person reading,
watching or operating something (a phone, a photograph, a paper,
any device) has its functional side — screen, front, page — facing
THEIR OWN eyes; the camera then sees whatever side the staging
geometry implies (often its back). Show the functional side to the
camera ONLY when the shot text itself stages it toward the viewer.
Never flip, mirror or reverse an object's front and back.

CARRIED STATE (persist exactly — must match the neighbouring shots of this scene): Su-ri-young’s bicycle remains hastily parked in the rear yard below, while the rooftop iron entrance door stands half open.

PEOPLE: the SHOT TEXT alone decides whether any person is visible in this shot. IF a person appears, they must be one of: 수리영 (한국인 20대 초반 여성, 긴 검은 생머리, 젊은 타원형 얼굴) — never anyone else, and never add a person the shot text does not show. IF only part of a person is in frame (a hand, arm, foot, back, silhouette), that body part belongs to the specific person the shot text names — its sex, age, build, skin and grooming must unmistakably match that person's profile above.

No text, captions, watermarks or annotations anywhere.

VISUAL APPROACH — this take (camera, framing, composition):
Photograph from close beside the half-open doorway, looking outward toward Su-ri-young in a frontal full-body composition. Use a compressed perspective so the edge of the metal door looms in the near foreground while her frozen stance and vacant, fixed gaze remain clearly readable beyond it. Shape the night light softly across her face while preserving darker falloff around her body, creating an intimate sense of shock and arrested movement.

PRIORITY OF INSTRUCTIONS: the VISUAL APPROACH above governs camera,
framing and composition ONLY. Every other instruction in this prompt —
location, time of day, people and their identity, poses, props,
carried state, physical rules, and the no-text rule — is binding
exactly as written earlier. If the VISUAL APPROACH appears to conflict
with any of them, the earlier instruction wins.

Edit this photograph. Apply ONLY the corrections listed below,
each in place, changing nothing else:

CORRECTIONS:
- Add the green cap to her head.
- Add the dark backpack to her back.

PRESERVE EVERYTHING ELSE EXACTLY as in the original photograph —
every person, pose, face, garment, object, set feature, lighting,
camera and framing stay identical. No text, captions or
watermarks anywhere.

Create ONE FINAL photorealistic live-action film still of the moment below — contemporary South Korea, 2026; all people are Korean unless stated. TIME OF DAY (lock): night.

SHOT TEXT (authoritative, Korean): 혜수가 멀어지는 수리영을 향해 입을 크게 벌리고 소리치는 정지 상태

LOCATION (lock): Beside the coastal bus-stop bench at night, facing the route along which the daughter runs away. The shot takes place here — the attached STORYBOARD SKETCH fixes the staging, camera and figure placement of this exact place. No location photograph is attached — build the location itself strictly from the location text above and the shot text, inventing nothing beyond them.

REALIZE FIGURATIVE LANGUAGE AS A LIVE-ACTION SHOT: the Korean shot
text may describe characters metaphorically, figuratively or with
exaggeration. Photograph what a real movie camera would actually
record on a physical set — exaggerated or figurative impressions
become realistic staging choices (lighting, distance, angle,
wardrobe), not literal fantasy imagery.
EVERY CHARACTER IS A HUMAN BEING: unless the story explicitly
features non-human or virtual beings (as in science fiction or
fantasy), every character — however indirectly, vaguely or
figuratively the text describes them — IS a real human. When the
text gives no direct visual description of a person, IMAGINE one
and still show them as a concrete, fully-formed human being:
always render the human form to the maximum extent — build,
posture, face, hands, clothing — never reduce a person to a shape,
blob, solid silhouette or abstract mass.

EXPRESSIONS ARE ACTED, NEVER ANATOMICAL: when the text describes a
person's eyes, face or presence emotionally or figuratively (vacant,
hollow, dazed, burning, lifeless gaze and the like), it describes an
ACTOR'S PERFORMANCE captured by a real camera — realize it ONLY
through gaze direction, focus, eyelids, facial muscles, stillness
and posture. Every living person's eyes remain anatomically normal
human eyes with a natural iris and pupil, natural sclera and normal
proportions; NEVER whiten, blank out, cloud over, glow, enlarge or
otherwise alter eyeballs, skin or anatomy — unless the story
explicitly declares that being non-human or supernatural in form.

PROPS FACE THE RIGHT WAY: every handheld or used object must be
oriented exactly as its real-world use requires. A person reading,
watching or operating something (a phone, a photograph, a paper,
any device) has its functional side — screen, front, page — facing
THEIR OWN eyes; the camera then sees whatever side the staging
geometry implies (often its back). Show the functional side to the
camera ONLY when the shot text itself stages it toward the viewer.
Never flip, mirror or reverse an object's front and back.

CARRIED STATE (persist exactly — must match the neighbouring shots of this scene): Hye-su still has her phone as Su-ri-young leaves with her own phone in her possession.

PEOPLE: the SHOT TEXT alone decides whether any person is visible in this shot. IF a person appears, they must be one of: 혜수 (한국인 40대 초반 여성, 단정한 검은 머리, 성숙한 타원형 얼굴) — never anyone else, and never add a person the shot text does not show. IF only part of a person is in frame (a hand, arm, foot, back, silhouette), that body part belongs to the specific person the shot text names — its sex, age, build, skin and grooming must unmistakably match that person's profile above.

No text, captions, watermarks or annotations anywhere.

Create ONE FINAL photorealistic live-action film still of the moment below — contemporary South Korea, 2026; all people are Korean unless stated. TIME OF DAY (lock): night.

SHOT TEXT (authoritative, Korean): 혜수가 멀어지는 수리영을 향해 입을 크게 벌리고 소리치는 정지 상태

LOCATION (lock): Beside the coastal bus-stop bench at night, facing the route along which the daughter runs away. The shot takes place here — the attached STORYBOARD SKETCH fixes the staging, camera and figure placement of this exact place. No location photograph is attached — build the location itself strictly from the location text above and the shot text, inventing nothing beyond them.

REALIZE FIGURATIVE LANGUAGE AS A LIVE-ACTION SHOT: the Korean shot
text may describe characters metaphorically, figuratively or with
exaggeration. Photograph what a real movie camera would actually
record on a physical set — exaggerated or figurative impressions
become realistic staging choices (lighting, distance, angle,
wardrobe), not literal fantasy imagery.
EVERY CHARACTER IS A HUMAN BEING: unless the story explicitly
features non-human or virtual beings (as in science fiction or
fantasy), every character — however indirectly, vaguely or
figuratively the text describes them — IS a real human. When the
text gives no direct visual description of a person, IMAGINE one
and still show them as a concrete, fully-formed human being:
always render the human form to the maximum extent — build,
posture, face, hands, clothing — never reduce a person to a shape,
blob, solid silhouette or abstract mass.

EXPRESSIONS ARE ACTED, NEVER ANATOMICAL: when the text describes a
person's eyes, face or presence emotionally or figuratively (vacant,
hollow, dazed, burning, lifeless gaze and the like), it describes an
ACTOR'S PERFORMANCE captured by a real camera — realize it ONLY
through gaze direction, focus, eyelids, facial muscles, stillness
and posture. Every living person's eyes remain anatomically normal
human eyes with a natural iris and pupil, natural sclera and normal
proportions; NEVER whiten, blank out, cloud over, glow, enlarge or
otherwise alter eyeballs, skin or anatomy — unless the story
explicitly declares that being non-human or supernatural in form.

PROPS FACE THE RIGHT WAY: every handheld or used object must be
oriented exactly as its real-world use requires. A person reading,
watching or operating something (a phone, a photograph, a paper,
any device) has its functional side — screen, front, page — facing
THEIR OWN eyes; the camera then sees whatever side the staging
geometry implies (often its back). Show the functional side to the
camera ONLY when the shot text itself stages it toward the viewer.
Never flip, mirror or reverse an object's front and back.

CARRIED STATE (persist exactly — must match the neighbouring shots of this scene): Hye-su still has her phone as Su-ri-young leaves with her own phone in her possession.

PEOPLE: the SHOT TEXT alone decides whether any person is visible in this shot. IF a person appears, they must be one of: 혜수 (한국인 40대 초반 여성, 단정한 검은 머리, 성숙한 타원형 얼굴) — never anyone else, and never add a person the shot text does not show. IF only part of a person is in frame (a hand, arm, foot, back, silhouette), that body part belongs to the specific person the shot text names — its sex, age, build, skin and grooming must unmistakably match that person's profile above.

No text, captions, watermarks or annotations anywhere.

VISUAL APPROACH — this take (camera, framing, composition):
Photograph the moment in a tight side-profile close-up at eye level, compressing the space so Hye-su’s mature oval face and wide-open shouting mouth dominate the frame. Keep her gaze driven sharply toward the offscreen route, with her phone retained naturally in hand near the lower edge of the crop. Use shallow focus and restrained, cool night illumination, letting the bus-stop surroundings and coastal route fall softly out of focus while preserving a realistic live-action texture.

PRIORITY OF INSTRUCTIONS: the VISUAL APPROACH above governs camera,
framing and composition ONLY. Every other instruction in this prompt —
location, time of day, people and their identity, poses, props,
carried state, physical rules, and the no-text rule — is binding
exactly as written earlier. If the VISUAL APPROACH appears to conflict
with any of them, the earlier instruction wins.

Create ONE FINAL photorealistic live-action film still of the moment below — contemporary South Korea, 2026; all people are Korean unless stated. TIME OF DAY (lock): night.

SHOT TEXT (authoritative, Korean): 혜수가 멀어지는 수리영을 향해 입을 크게 벌리고 소리치는 정지 상태

LOCATION (lock): Beside the coastal bus-stop bench at night, facing the route along which the daughter runs away. The shot takes place here — the attached STORYBOARD SKETCH fixes the staging, camera and figure placement of this exact place. No location photograph is attached — build the location itself strictly from the location text above and the shot text, inventing nothing beyond them.

REALIZE FIGURATIVE LANGUAGE AS A LIVE-ACTION SHOT: the Korean shot
text may describe characters metaphorically, figuratively or with
exaggeration. Photograph what a real movie camera would actually
record on a physical set — exaggerated or figurative impressions
become realistic staging choices (lighting, distance, angle,
wardrobe), not literal fantasy imagery.
EVERY CHARACTER IS A HUMAN BEING: unless the story explicitly
features non-human or virtual beings (as in science fiction or
fantasy), every character — however indirectly, vaguely or
figuratively the text describes them — IS a real human. When the
text gives no direct visual description of a person, IMAGINE one
and still show them as a concrete, fully-formed human being:
always render the human form to the maximum extent — build,
posture, face, hands, clothing — never reduce a person to a shape,
blob, solid silhouette or abstract mass.

EXPRESSIONS ARE ACTED, NEVER ANATOMICAL: when the text describes a
person's eyes, face or presence emotionally or figuratively (vacant,
hollow, dazed, burning, lifeless gaze and the like), it describes an
ACTOR'S PERFORMANCE captured by a real camera — realize it ONLY
through gaze direction, focus, eyelids, facial muscles, stillness
and posture. Every living person's eyes remain anatomically normal
human eyes with a natural iris and pupil, natural sclera and normal
proportions; NEVER whiten, blank out, cloud over, glow, enlarge or
otherwise alter eyeballs, skin or anatomy — unless the story
explicitly declares that being non-human or supernatural in form.

PROPS FACE THE RIGHT WAY: every handheld or used object must be
oriented exactly as its real-world use requires. A person reading,
watching or operating something (a phone, a photograph, a paper,
any device) has its functional side — screen, front, page — facing
THEIR OWN eyes; the camera then sees whatever side the staging
geometry implies (often its back). Show the functional side to the
camera ONLY when the shot text itself stages it toward the viewer.
Never flip, mirror or reverse an object's front and back.

CARRIED STATE (persist exactly — must match the neighbouring shots of this scene): Hye-su still has her phone as Su-ri-young leaves with her own phone in her possession.

PEOPLE: the SHOT TEXT alone decides whether any person is visible in this shot. IF a person appears, they must be one of: 혜수 (한국인 40대 초반 여성, 단정한 검은 머리, 성숙한 타원형 얼굴) — never anyone else, and never add a person the shot text does not show. IF only part of a person is in frame (a hand, arm, foot, back, silhouette), that body part belongs to the specific person the shot text names — its sex, age, build, skin and grooming must unmistakably match that person's profile above.

No text, captions, watermarks or annotations anywhere.

VISUAL APPROACH — this take (camera, framing, composition):
Use a wide, static view from around chest height, holding Hye-su beside the coastal bus-stop bench while giving strong compositional weight to the empty route she faces. Place her off-center, her open-mouthed shout and forward-directed posture clearly readable against the deeper night setting, with her phone visibly still in her possession. Maintain broad depth and subdued naturalistic night contrast so the distance ahead feels palpable and her isolation within the frame carries the dramatic beat.

PRIORITY OF INSTRUCTIONS: the VISUAL APPROACH above governs camera,
framing and composition ONLY. Every other instruction in this prompt —
location, time of day, people and their identity, poses, props,
carried state, physical rules, and the no-text rule — is binding
exactly as written earlier. If the VISUAL APPROACH appears to conflict
with any of them, the earlier instruction wins.

Edit this photograph. Apply ONLY the corrections listed below,
each in place, changing nothing else:

CORRECTIONS:
- Correct the anatomy of the hand holding the phone so the fingers and grip are naturally and realistically formed.

PRESERVE EVERYTHING ELSE EXACTLY as in the original photograph —
every person, pose, face, garment, object, set feature, lighting,
camera and framing stay identical. No text, captions or
watermarks anywhere.

Create ONE FINAL photorealistic live-action film still of the moment below — contemporary South Korea, 2026; all people are Korean unless stated. TIME OF DAY (lock): dusk.

SHOT TEXT (authoritative, Korean): 황금빛 노을이 반사되는 바다 위에 소형 어선 한 척이 떠 있는 넓은 전경

LOCATION (lock): On open water at sunset, showing the small fishing boat surrounded by a golden reflection on the sea. The shot takes place here — the attached STORYBOARD SKETCH fixes the staging, camera and figure placement of this exact place. No location photograph is attached — build the location itself strictly from the location text above and the shot text, inventing nothing beyond them.

REALIZE FIGURATIVE LANGUAGE AS A LIVE-ACTION SHOT: the Korean shot
text may describe characters metaphorically, figuratively or with
exaggeration. Photograph what a real movie camera would actually
record on a physical set — exaggerated or figurative impressions
become realistic staging choices (lighting, distance, angle,
wardrobe), not literal fantasy imagery.
EVERY CHARACTER IS A HUMAN BEING: unless the story explicitly
features non-human or virtual beings (as in science fiction or
fantasy), every character — however indirectly, vaguely or
figuratively the text describes them — IS a real human. When the
text gives no direct visual description of a person, IMAGINE one
and still show them as a concrete, fully-formed human being:
always render the human form to the maximum extent — build,
posture, face, hands, clothing — never reduce a person to a shape,
blob, solid silhouette or abstract mass.

EXPRESSIONS ARE ACTED, NEVER ANATOMICAL: when the text describes a
person's eyes, face or presence emotionally or figuratively (vacant,
hollow, dazed, burning, lifeless gaze and the like), it describes an
ACTOR'S PERFORMANCE captured by a real camera — realize it ONLY
through gaze direction, focus, eyelids, facial muscles, stillness
and posture. Every living person's eyes remain anatomically normal
human eyes with a natural iris and pupil, natural sclera and normal
proportions; NEVER whiten, blank out, cloud over, glow, enlarge or
otherwise alter eyeballs, skin or anatomy — unless the story
explicitly declares that being non-human or supernatural in form.

PROPS FACE THE RIGHT WAY: every handheld or used object must be
oriented exactly as its real-world use requires. A person reading,
watching or operating something (a phone, a photograph, a paper,
any device) has its functional side — screen, front, page — facing
THEIR OWN eyes; the camera then sees whatever side the staging
geometry implies (often its back). Show the functional side to the
camera ONLY when the shot text itself stages it toward the viewer.
Never flip, mirror or reverse an object's front and back.

CARRIED STATE (persist exactly — must match the neighbouring shots of this scene): Su-ri-young and In-woo are aboard the same small fishing boat. Her backpack, printed maps, old photograph, coordinate note, and phone are all aboard with her.

NO PEOPLE IN THIS SHOT: the shot text shows only the place and its state — no living person or any body part appears in frame, unless the shot text itself explicitly says so.

No text, captions, watermarks or annotations anywhere.

Create ONE FINAL photorealistic live-action film still of the moment below — contemporary South Korea, 2026; all people are Korean unless stated. TIME OF DAY (lock): dusk.

SHOT TEXT (authoritative, Korean): 황금빛 노을이 반사되는 바다 위에 소형 어선 한 척이 떠 있는 넓은 전경

LOCATION (lock): On open water at sunset, showing the small fishing boat surrounded by a golden reflection on the sea. The shot takes place here — the attached STORYBOARD SKETCH fixes the staging, camera and figure placement of this exact place. No location photograph is attached — build the location itself strictly from the location text above and the shot text, inventing nothing beyond them.

REALIZE FIGURATIVE LANGUAGE AS A LIVE-ACTION SHOT: the Korean shot
text may describe characters metaphorically, figuratively or with
exaggeration. Photograph what a real movie camera would actually
record on a physical set — exaggerated or figurative impressions
become realistic staging choices (lighting, distance, angle,
wardrobe), not literal fantasy imagery.
EVERY CHARACTER IS A HUMAN BEING: unless the story explicitly
features non-human or virtual beings (as in science fiction or
fantasy), every character — however indirectly, vaguely or
figuratively the text describes them — IS a real human. When the
text gives no direct visual description of a person, IMAGINE one
and still show them as a concrete, fully-formed human being:
always render the human form to the maximum extent — build,
posture, face, hands, clothing — never reduce a person to a shape,
blob, solid silhouette or abstract mass.

EXPRESSIONS ARE ACTED, NEVER ANATOMICAL: when the text describes a
person's eyes, face or presence emotionally or figuratively (vacant,
hollow, dazed, burning, lifeless gaze and the like), it describes an
ACTOR'S PERFORMANCE captured by a real camera — realize it ONLY
through gaze direction, focus, eyelids, facial muscles, stillness
and posture. Every living person's eyes remain anatomically normal
human eyes with a natural iris and pupil, natural sclera and normal
proportions; NEVER whiten, blank out, cloud over, glow, enlarge or
otherwise alter eyeballs, skin or anatomy — unless the story
explicitly declares that being non-human or supernatural in form.

PROPS FACE THE RIGHT WAY: every handheld or used object must be
oriented exactly as its real-world use requires. A person reading,
watching or operating something (a phone, a photograph, a paper,
any device) has its functional side — screen, front, page — facing
THEIR OWN eyes; the camera then sees whatever side the staging
geometry implies (often its back). Show the functional side to the
camera ONLY when the shot text itself stages it toward the viewer.
Never flip, mirror or reverse an object's front and back.

CARRIED STATE (persist exactly — must match the neighbouring shots of this scene): Su-ri-young and In-woo are aboard the same small fishing boat. Her backpack, printed maps, old photograph, coordinate note, and phone are all aboard with her.

NO PEOPLE IN THIS SHOT: the shot text shows only the place and its state — no living person or any body part appears in frame, unless the shot text itself explicitly says so.

No text, captions, watermarks or annotations anywhere.

VISUAL APPROACH — this take (camera, framing, composition):
Photograph from a high, very distant vantage in an expansive establishing view, letting the golden sunset reflection dominate the open sea while the fishing boat remains a small, clearly readable subject within the broad composition. Use a deep, spacious perspective and restrained dusk contrast, with the reflected light forming the primary visual path toward the boat.

PRIORITY OF INSTRUCTIONS: the VISUAL APPROACH above governs camera,
framing and composition ONLY. Every other instruction in this prompt —
location, time of day, people and their identity, poses, props,
carried state, physical rules, and the no-text rule — is binding
exactly as written earlier. If the VISUAL APPROACH appears to conflict
with any of them, the earlier instruction wins.

Create ONE FINAL photorealistic live-action film still of the moment below — contemporary South Korea, 2026; all people are Korean unless stated. TIME OF DAY (lock): dusk.

SHOT TEXT (authoritative, Korean): 황금빛 노을이 반사되는 바다 위에 소형 어선 한 척이 떠 있는 넓은 전경

LOCATION (lock): On open water at sunset, showing the small fishing boat surrounded by a golden reflection on the sea. The shot takes place here — the attached STORYBOARD SKETCH fixes the staging, camera and figure placement of this exact place. No location photograph is attached — build the location itself strictly from the location text above and the shot text, inventing nothing beyond them.

REALIZE FIGURATIVE LANGUAGE AS A LIVE-ACTION SHOT: the Korean shot
text may describe characters metaphorically, figuratively or with
exaggeration. Photograph what a real movie camera would actually
record on a physical set — exaggerated or figurative impressions
become realistic staging choices (lighting, distance, angle,
wardrobe), not literal fantasy imagery.
EVERY CHARACTER IS A HUMAN BEING: unless the story explicitly
features non-human or virtual beings (as in science fiction or
fantasy), every character — however indirectly, vaguely or
figuratively the text describes them — IS a real human. When the
text gives no direct visual description of a person, IMAGINE one
and still show them as a concrete, fully-formed human being:
always render the human form to the maximum extent — build,
posture, face, hands, clothing — never reduce a person to a shape,
blob, solid silhouette or abstract mass.

EXPRESSIONS ARE ACTED, NEVER ANATOMICAL: when the text describes a
person's eyes, face or presence emotionally or figuratively (vacant,
hollow, dazed, burning, lifeless gaze and the like), it describes an
ACTOR'S PERFORMANCE captured by a real camera — realize it ONLY
through gaze direction, focus, eyelids, facial muscles, stillness
and posture. Every living person's eyes remain anatomically normal
human eyes with a natural iris and pupil, natural sclera and normal
proportions; NEVER whiten, blank out, cloud over, glow, enlarge or
otherwise alter eyeballs, skin or anatomy — unless the story
explicitly declares that being non-human or supernatural in form.

PROPS FACE THE RIGHT WAY: every handheld or used object must be
oriented exactly as its real-world use requires. A person reading,
watching or operating something (a phone, a photograph, a paper,
any device) has its functional side — screen, front, page — facing
THEIR OWN eyes; the camera then sees whatever side the staging
geometry implies (often its back). Show the functional side to the
camera ONLY when the shot text itself stages it toward the viewer.
Never flip, mirror or reverse an object's front and back.

CARRIED STATE (persist exactly — must match the neighbouring shots of this scene): Su-ri-young and In-woo are aboard the same small fishing boat. Her backpack, printed maps, old photograph, coordinate note, and phone are all aboard with her.

NO PEOPLE IN THIS SHOT: the shot text shows only the place and its state — no living person or any body part appears in frame, unless the shot text itself explicitly says so.

No text, captions, watermarks or annotations anywhere.

VISUAL APPROACH — this take (camera, framing, composition):
Photograph from a low viewpoint close to water level with a compressed long-lens feel, isolating the fishing boat against the dense golden reflection across the sea. Keep the framing wide enough to preserve the scale of the open water, but make the layered bands of reflected dusk light and the boat’s profile feel graphic, distant, and quietly suspended.

PRIORITY OF INSTRUCTIONS: the VISUAL APPROACH above governs camera,
framing and composition ONLY. Every other instruction in this prompt —
location, time of day, people and their identity, poses, props,
carried state, physical rules, and the no-text rule — is binding
exactly as written earlier. If the VISUAL APPROACH appears to conflict
with any of them, the earlier instruction wins.

Edit this photograph. Apply ONLY the corrections listed below,
each in place, changing nothing else:

CORRECTIONS:
- Remove the dark silhouettes of people from the deck of the boat so it appears completely empty.

PRESERVE EVERYTHING ELSE EXACTLY as in the original photograph —
every person, pose, face, garment, object, set feature, lighting,
camera and framing stay identical. No text, captions or
watermarks anywhere.