{ "status": "completed", "applicable_count": 4, "completed_count": 4, "failed_count": 0, "schema_version": 1, "config_hash": "78e79f6fbb05a6af", "data": { "schema_version": 1, "groups": [ { "group_id": "vca-s3-zoom-1", "anchor_type": "zoom_continuity", "members": [ { "scene_index": 3, "shot_index": 1, "role": "source_wide" }, { "scene_index": 3, "shot_index": 6, "role": "zoom" } ], "locked_elements": [ { "confidence": "high", "description": "혜수 stands among the detectives as the officer directing the investigation.", "evidence_quote": "형사무리 속에 사건을 지휘하는 혜수의 모습이 보인다.", "kind": "subject_pose" }, { "confidence": "high", "description": "혜수 is nearly frontal with a resolute face and a braced command posture.", "evidence_quote": "Keep her nearly frontal as a deliberate image of command", "kind": "subject_pose" }, { "confidence": "high", "description": "혜수’s eyes remain fixed just beyond the lens rather than looking at the audience.", "evidence_quote": "her eyes remain fixed just beyond the lens rather than addressing the audience.", "kind": "subject_pose" }, { "confidence": "high", "description": "The detective group is clustered behind 혜수, establishing her as its central figure.", "evidence_quote": "with the detective group compressed softly behind her shoulders.", "kind": "subject_pose" }, { "confidence": "high", "description": "The police line remains installed as a lateral boundary at the investigation scene.", "evidence_quote": "Its length runs laterally across the frontal camera view.", "kind": "prop_state" }, { "confidence": "high", "description": "Both views are presented as local-news footage of the investigation.", "evidence_quote": "지역 뉴스 화면.", "kind": "surface_state" } ], "wide_shot_contract": [ "The wide frame must visibly contain 혜수 standing among the detective group and directing the scene.", "혜수 must be visible from a compatible nearly frontal side, with her eyes directed just beyond the lens.", "The detective group must occupy the area behind or around 혜수 so the close framing can isolate her with detectives behind her shoulders.", "The installed police line must remain visible as a lateral boundary within the investigation scene.", "The entire view must retain the appearance of local-news investigation footage." ], "camera_relation": "same_axis_punch_in", "camera_relation_evidence": "Wide staging: “hold an eye-level wide frontal view from outside the cordon” and “leaves a clear visual path for the subsequent advance toward ground-level investigators.” Zoom staging: “The news camera completes its eye-level frontal dolly-in until 혜수's resolute face occupies most of the frame” and “Keep her nearly frontal.”", "per_shot_consumption_hints": { "3:1": { "include_in_wide": true }, "3:6": { "crop_from_source": true } }, "revised_wide_t2i": { "0": "Photorealistic cinematic still. At eye level from outside the cordon, a continuous installed police line spans laterally across the lower foreground as a clear boundary, with a forested mountain trail extending through the center and three East Asian police personnel distributed across the middle distance inside the investigation area. One uniformed officer occupies the left side, while the plainclothes investigator 혜수 stands farther along the same trail among the investigative team as the officer visibly directing the scene, and another investigator holds a low kneeling posture near the ground on the right. 혜수 is presented from a compatible nearly frontal side with a resolute expression and braced command posture, her eyes fixed just beyond the lens rather than addressing the audience; the left officer and right kneeling investigator remain clearly separated in the wide composition while occupying the depth behind and around 혜수, establishing her as the central figure of the detective group. None looks directly toward the camera. The camera focus emphasizes the foreground police line and the clearly separated investigators beyond it, preserving an unobstructed visual path along the trail. Natural daylight, restrained contrast, muted color, subdued local-television-news investigation-footage texture, slight digital compression, no visible device border or screen frame." }, "revised_wide_prompt_provenance": { "0": { "original_scene_detail_prompt_hash": "90482b9bd31b1bb7", "revised_prompt_hash": "bcdf67390efdc359" } } }, { "group_id": "vca-s7-zoom-2", "anchor_type": "zoom_continuity", "members": [ { "scene_index": 7, "shot_index": 10, "role": "source_wide" }, { "scene_index": 7, "shot_index": 14, "role": "zoom" } ], "locked_elements": [ { "confidence": "high", "description": "수리영 must be holding a smartphone below her chin and looking down at it.", "evidence_quote": "\"body_pose\": \"phone held below chin\" ... \"gaze_direction_kind\": \"down\"", "kind": "held_prop" }, { "confidence": "high", "description": "The smartphone must display an unanswered-call screen.", "evidence_quote": "\"The front display faces the camera obliquely and shows an unanswered call.\"", "kind": "surface_state" }, { "confidence": "high", "description": "수리영’s expression must remain fixed and hardened as she checks the unanswered call.", "evidence_quote": "응답 없는 통화 화면을 내려다보는 수리영의 굳은 표정.", "kind": "subject_pose" } ], "wide_shot_contract": [ "The wide frame must include 수리영 holding the smartphone below her chin and looking down at it.", "The smartphone in the wide frame must be the same phone displaying the unanswered call, even if the screen is too small to read at wide scale.", "수리영’s fixed, hardened expression must already be present in the wide shot." ], "camera_relation": "different_camera_same_moment", "camera_relation_evidence": "WIDE: \"Track laterally at 수리영’s eye level from a position perpendicular to her sightline, preserving a clean profile in the left third\" and \"angle\": \"profile_right\". ZOOM: \"Dolly in from just below 수리영’s eye line, placing the phone in the lower foreground at a readable angle while her face remains in the upper half\" and \"angle\": \"three_quarter_right\". A clean profile crop cannot produce the zoom’s three-quarter facial view and oblique view of the phone display.", "per_shot_consumption_hints": { "7:10": { "include_in_wide": true }, "7:14": { "crop_from_source": false } }, "revised_wide_t2i": { "0": "Photorealistic cinematic still: C01O01 in a gray hoodie, a young Korean woman with long black hair, an oval face, and distinct features, occupies the middle-left midground in right profile beside the occupied dog enclosures in the middle-right background. An eye-level observer camera perpendicular to her sightline preserves her clean profile in the left third, with her upper body inclined toward the existing kennel enclosures across the right half. She holds a smartphone below her chin and looks down at it with a fixed, hardened expression; the phone’s front display faces the camera obliquely and visibly shows an unanswered-call screen, even though its details are small at this wide scale. Her lips are slightly parted and her shoulders are subtly lowered. Focus holds on her profile, the phone, and several confined dogs already present within the established background reference. The enclosure fronts face partly toward the camera and extend behind her profile within the unchanged shelter setting. Restrained daytime ambient light maintains muted natural color and neutral low-to-mid-key contrast without a dramatic color cast." }, "revised_wide_prompt_provenance": { "0": { "original_scene_detail_prompt_hash": "17723ad816d6a9e3", "revised_prompt_hash": "c4ce6beed167f30e" } }, "revised_zoom_t2i": { "0": "Photorealistic cinematic still. From an observer position just below eye level, C01O01 in a gray hoodie, a young Korean woman, occupies the upper half at three-quarter angle from the right, her long black hair framing an oval face with clearly defined features, her upright posture and visibly rigid, fixed and hardened expression held as she gazes downward at the smartphone she holds below her chin. The smartphone sits in the lower foreground, proportionally smaller than her face; its front display faces the camera obliquely, showing a legible unanswered-call screen. Balanced shallow focus favors her facial tension without obscuring the display. Neutral, restrained daytime ambient light reveals both surfaces in muted natural color without a dramatic cast. Only indistinct portions of the shelter enclosure fronts remain softly visible behind her." }, "revised_zoom_prompt_provenance": { "0": { "original_scene_detail_prompt_hash": "2cebcc0bd65d6f7e", "revised_prompt_hash": "781c761a9c0e8d56" } } }, { "group_id": "vca-s20-zoom-3", "anchor_type": "zoom_continuity", "members": [ { "scene_index": 20, "shot_index": 1, "role": "source_wide" }, { "scene_index": 20, "shot_index": 6, "role": "zoom" } ], "locked_elements": [ { "confidence": "high", "description": "The CCTV monitor must be playing the same specific section of footage that is exceptionally dark, crushed, and too low-detail for identification.", "evidence_quote": "“해당 구간을 재생 중이며 인물 식별이 불가능할 만큼 어둡게 뭉개짐”", "kind": "surface_state" }, { "confidence": "high", "description": "Two apparent human figures must remain present as indistinct forms within the dark CCTV footage, with neither figure identifiable.", "evidence_quote": "“The unusually dark section of footage becomes an abstracted close field in which the two apparent figures cannot be identified.”", "kind": "subject_pose" } ], "wide_shot_contract": [ "The wide shot must visibly include the CCTV monitor and its image-bearing playback face.", "The monitor must visibly show the same exceptionally dark, low-detail playback section containing two indistinct apparent human figures." ], "camera_relation": "different_camera_same_moment", "camera_relation_evidence": "WIDE: “The playback face is angled toward the seated pair and seen obliquely by the camera, displaying the reviewed CCTV footage.” ZOOM: “Directly in front of the CCTV monitor at seated eye level” and “The image-bearing front faces the camera squarely.” A crop of the oblique wide view cannot produce the zoom’s square-on presentation of the display.", "per_shot_consumption_hints": { "20:1": { "include_in_wide": true }, "20:6": { "crop_from_source": false } }, "revised_wide_t2i": { "0": "Photorealistic cinematic still. From standing eye level behind and slightly beside C01O01 in a gray hoodie, a young East Asian woman with long black hair stands with back to camera at the left foreground edge, hovering behind the seated pair and looking down toward the CCTV display. At the same workstation in the midground, an adult security employee in a security uniform sits in left profile with dark hair, ordinary facial contours, neat features, ethnicity and nationality visually indiscernible, one hand resting on the controls and gaze directed down. C02O02 in a dark navy blouson, a middle-aged East Asian woman with dark hair, a mature oval face, and neat features, remains seated in the existing chair at three-quarter angle from the right, leaning forward with gaze directed down toward the monitor. The existing desk supports the shared workstation and places the two seated viewers and monitor in a triangular arrangement beneath C01O01's shoulder line. The CCTV monitor and its visible, image-bearing playback face are included in the wide frame, angled toward the seated pair and seen obliquely by the camera; it is playing the same specific section of footage that is exceptionally dark, crushed, and too low-detail for identification, with heavily muted tonal separation and two apparent human figures remaining visible only as indistinct forms, neither identifiable by facial or bodily detail. Focus rests on the oblique playback face, the seated pair's downward sightlines, and C01O01's rigid foreground posture within the same shared workstation space. The attached L16B01 background reference remains unchanged. Subdued neutral interior ambient illumination surrounds the workstation with restrained contrast, no strong color cast, and no dramatic directional light." }, "revised_wide_prompt_provenance": { "0": { "original_scene_detail_prompt_hash": "58612af50b2f2e5e", "revised_prompt_hash": "262fcd06525f2bb5" } }, "revised_zoom_t2i": { "0": "Photorealistic cinematic still. The existing CCTV monitor from the attached background reference faces the camera squarely at seated eye level, its displayed image filling nearly the entire frame, with only a narrow trace of the monitor's front rim at the edge. The monitor is playing the same specific section of footage, exceptionally dark, crushed, and too low-detail for identification. Within this indistinct CCTV space, two apparent adult human figures remain present as vague forms oriented toward each other, with neither figure identifiable and no discernible facial features, bodily detail, clothing, or identity cues. Heavily muted tonal separation reduces the image to an abstract low-detail field with slight digital texture and reduced resolution. No viewers or surrounding room furnishings are visible." }, "revised_zoom_prompt_provenance": { "0": { "original_scene_detail_prompt_hash": "75030112843e30c0", "revised_prompt_hash": "35664ad272e8e1ca" } } }, { "group_id": "vca-s25-zoom-4", "anchor_type": "zoom_continuity", "members": [ { "scene_index": 25, "shot_index": 8, "role": "source_wide" }, { "scene_index": 25, "shot_index": 17, "role": "zoom" } ], "locked_elements": [ { "confidence": "high", "description": "The same old, blood-stained photograph is shown front-facing in both shots.", "evidence_quote": "“Its image-bearing face points toward the camera from the exact center of the red circle.” / “Its image-bearing front face is viewed straight on” / “Old and blood-stained.”", "kind": "prop_state" }, { "confidence": "high", "description": "The photograph depicts two twelve-year-old girls seated together in front of a rural restaurant.", "evidence_quote": "“식당 앞에는 어린 여자애 둘(12살)이 앉아있다.”", "kind": "subject_pose" }, { "confidence": "high", "description": "Young Kang Min-sook sits at a three-quarter-left angle beside the other girl and looks off-screen.", "evidence_quote": "“angle”: “three_quarter_left”, “body_pose”: “perched beside companion”, “character”: “어린 강민숙”, “gaze_direction_kind”: “off_screen”", "kind": "subject_pose" }, { "confidence": "high", "description": "The second twelve-year-old girl is in right profile with her knees drawn close and her gaze lowered.", "evidence_quote": "“angle”: “profile_right”, “body_pose”: “knees drawn close”, “character”: “열두 살 여자아이”, “gaze_direction_kind”: “down”", "kind": "subject_pose" }, { "confidence": "high", "description": "The depicted restaurant has a round installation mounted on its roof and visible above the roofline.", "evidence_quote": "“지붕위에 둥근 원이 설치된 (마치 십자가처럼)”", "kind": "prop_state" }, { "confidence": "high", "description": "The restaurant is depicted as a rural restaurant from approximately the 1970s–1980s, with its front facade facing the photographed viewpoint.", "evidence_quote": "“7-80년대 쯤 보이는 시골 식당” / “The front facade faces the viewpoint within the photograph.”", "kind": "prop_state" } ], "wide_shot_contract": [ "The wide shot must visibly contain the blood-stained photograph fixed at the exact center of the complete red circle on the bedroom wall.", "The photograph’s image-bearing face must be square to the camera and sufficiently legible to contain the full zoom composition.", "Within the photograph, two twelve-year-old girls must sit side by side across the lower portion in front of the rural restaurant: young Kang Min-sook at a three-quarter-left angle looking off-screen, and the other girl in right profile with knees drawn close and looking down.", "The photographed restaurant’s front facade and the round rooftop installation must remain visible above the girls." ], "camera_relation": "same_axis_punch_in", "camera_relation_evidence": "WIDE: “Aligned squarely with the bedroom wall, the dolly-in centers the blood-stained photograph inside the painted red circle” and “Its image-bearing face points toward the camera from the exact center of the red circle.” ZOOM: “The cut lands orthogonal to the old photograph's image plane” and “Its image-bearing front face is viewed straight on.”", "per_shot_consumption_hints": { "25:8": { "include_in_wide": true }, "25:17": { "crop_from_source": true } }, "revised_wide_t2i": { "0": "Photorealistic cinematic still. P08, the same old blood-stained photograph, occupies the middle-center midground, fixed at the exact center of the complete deep-red circle on the bedroom wall in the middle-center background. The marked wall faces the camera squarely, the entire deep-red circle visible around the smaller photograph, with a narrow perimeter of the wall remaining inside the frame. P08 is a small handheld rectangular photographic print whose image-bearing face is directed squarely toward the camera and remains sufficiently legible to contain the full depicted composition. The worn glossy print has faded yellow-green color, softened corners, light scratches, small stains, crumpling creases, and visible bloodstaining. Its printed image shows the front-facing facade of an old rural Korean restaurant dating from approximately the 1970s–1980s, with a small area of writing and a rounded ring-like installation mounted on the roof and clearly visible above the roofline. Across the lower portion of the printed scene, two girls around twelve sit side by side in front of the restaurant: young Kang Min-sook, a Korean girl with a round youthful face and straight dark hair, sits at a three-quarter-left angle beside her companion and looks off-screen; the second girl, of indeterminate ethnicity, is shown in right profile with her knees drawn close and her gaze lowered. The restaurant facade and round rooftop installation remain visible above both girls. Camera focus rests sharply on the printed scene and blood marks. Fading natural light from dusk covers the wall with subdued, restrained contrast, leaving the deep-red circle and blood accents selectively distinct against the dim neutral wall." }, "revised_wide_prompt_provenance": { "0": { "original_scene_detail_prompt_hash": "e5a9cced63ab9f17", "revised_prompt_hash": "9f4b5898a7ba5eaa" } } } ], "diagnostics": { "seeds_detected": 8, "groups_extracted": 4, "skipped": [ { "seed": "S19sh12<-S8sh8", "reason": "cross_scene_zoom" }, { "seed": "S21sh29<-S16sh15", "reason": "cross_scene_zoom" }, { "seed": "S25sh8<-S12sh42", "reason": "cross_scene_zoom" }, { "seed": "S27sh4<-S25sh27", "reason": "cross_scene_zoom" }, { "seed": "S13sh13", "reason": "zoom_subject_absent_in_source", "missing_chars": [ "C02" ] }, { "seed": "S19sh22", "reason": "zoom_subject_absent_in_source", "missing_chars": [ "C01" ] }, { "seed": "S26sh28", "reason": "zoom_subject_absent_in_source", "missing_chars": [ "C03" ] }, { "seed": "S26sh32", "reason": "zoom_subject_absent_in_source", "missing_chars": [ "C01" ] } ], "revised_prompt_audit_violations": [], "rewrite_variation_skipped": [], "candidate_same_moment_exact_background": [ { "zoom": [ 1, 7 ], "source": [ 1, 3 ] }, { "zoom": [ 5, 24 ], "source": [ 5, 21 ] }, { "zoom": [ 14, 28 ], "source": [ 14, 12 ] }, { "zoom": [ 16, 15 ], "source": [ 16, 12 ] }, { "zoom": [ 21, 44 ], "source": [ 21, 29 ] } ], "ref_usage_source": "shot_dependency_t2i" }, "review_html": "review_html/index.html" }, "step_id": "zoom_continuity_anchor", "run_id": "ed5ce645-4811-4d60-8e72-5529dc7ab819", "resolved_model": "gpt", "updated_at": "2026-07-10T11:55:24.178932+00:00", "project_config_snapshot": {} }