{ "status": "completed", "applicable_count": 38, "completed_count": 38, "failed_count": 0, "schema_version": 1, "config_hash": "fb5bd23a76860bca", "data": { "contis": { "S5sh1": { "image_path": "/Users/manta/Documents/Projects/TheRoad-I1/projects/ecedba66-2474-4dff-a1a4-7d13bac08e1d/images/5ea07552-480a-4fdd-a888-b6147ff239f5/conti/conti_S5sh1.png", "plate_path": "/Users/manta/Documents/Projects/TheRoad-I1/projects/ecedba66-2474-4dff-a1a4-7d13bac08e1d/episodes/5ea07552-480a-4fdd-a888-b6147ff239f5/images/background_chain/L04B02.png", "prompt": "Draw ONE storyboard layout frame, 16:9, filling the entire image edge-to-edge.\n\nSTYLE — LAYOUT-ONLY THUMBNAIL SKETCH: the lightest possible\ndirector's layout drawing. THIN, CLEAN, UNIFORM PENCIL CONTOUR\nLINES ONLY on pure white paper. Absolutely NO shading, NO tone, NO\nhatching, NO texture, NO grayscale fill, NO light-and-shadow, NO\nfacial features beyond a bare head outline, NO clothing detail, NO\nmaterial rendering. The frame contains ONLY: the camera framing,\neach figure as a simple correct-proportion human contour in its\nexact pose and screen position/size, and the few major set masses\nas plain outline shapes. It must read as a quick layout guide,\nnever as a finished drawing.\n\nThe attached photograph is the LOCATION where this shot happens.\nUse it ONLY as a LOOSE spatial reference — the rough set masses\nand where things sit relative to each other. You do NOT need to\nfollow it exactly: do not copy its camera, its lighting or its\ndetails. Choose the camera and framing the SHOT TEXT needs, and\ndraw only the few set masses that matter as plain outline shapes.\n\nTHE LOCATION: 옥탑방 내부 주거 공간\n\nSHOT TEXT (authoritative, Korean): 작은 창문으로 들어오는 가느다란 햇살 아래, 앞치마를 두른 강민숙(민숙)이 굳은 표정으로 거실 TV 화면에 시선을 고정한 모습.\n\nCARRIED STATE (persist exactly): Minsuk already bears the small circular mark on her wrist; it is a long-standing physical feature even when not shown in this framing.\n\nABSOLUTELY NO TEXT anywhere: no captions, no numbers, no labels of any kind.", "skipped_reason": null, "error": null, "asset_id": "405d7f2d-b423-4a3f-9f17-436003468c56" }, "S5sh7": { "image_path": "/Users/manta/Documents/Projects/TheRoad-I1/projects/ecedba66-2474-4dff-a1a4-7d13bac08e1d/images/5ea07552-480a-4fdd-a888-b6147ff239f5/conti/conti_S5sh7.png", "plate_path": "/Users/manta/Documents/Projects/TheRoad-I1/projects/ecedba66-2474-4dff-a1a4-7d13bac08e1d/episodes/5ea07552-480a-4fdd-a888-b6147ff239f5/images/background_chain/L04B02.png", "prompt": "Draw ONE storyboard layout frame, 16:9, filling the entire image edge-to-edge.\n\nSTYLE — LAYOUT-ONLY THUMBNAIL SKETCH: the lightest possible\ndirector's layout drawing. THIN, CLEAN, UNIFORM PENCIL CONTOUR\nLINES ONLY on pure white paper. Absolutely NO shading, NO tone, NO\nhatching, NO texture, NO grayscale fill, NO light-and-shadow, NO\nfacial features beyond a bare head outline, NO clothing detail, NO\nmaterial rendering. The frame contains ONLY: the camera framing,\neach figure as a simple correct-proportion human contour in its\nexact pose and screen position/size, and the few major set masses\nas plain outline shapes. It must read as a quick layout guide,\nnever as a finished drawing.\n\nThe attached photograph is the LOCATION where this shot happens.\nUse it ONLY as a LOOSE spatial reference — the rough set masses\nand where things sit relative to each other. You do NOT need to\nfollow it exactly: do not copy its camera, its lighting or its\ndetails. Choose the camera and framing the SHOT TEXT needs, and\ndraw only the few set masses that matter as plain outline shapes.\n\nTHE LOCATION: 옥탑방 내부 주거 공간\n\nSHOT TEXT (authoritative, Korean): 수리영이 강민숙(민숙)의 어깨 가까이에 코를 댄 채 옅은 미소를 짓는 옆얼굴.\n\nCARRIED STATE (persist exactly): The small circular mark remains on Minsuk's wrist while Suriyoung leans close to her.\n\nABSOLUTELY NO TEXT anywhere: no captions, no numbers, no labels of any kind.", "skipped_reason": null, "error": null, "asset_id": "4facbb8d-821a-4825-9a78-47ba1828eccc" }, "S5sh8": { "image_path": "/Users/manta/Documents/Projects/TheRoad-I1/projects/ecedba66-2474-4dff-a1a4-7d13bac08e1d/images/5ea07552-480a-4fdd-a888-b6147ff239f5/conti/conti_S5sh8.png", "plate_path": "/Users/manta/Documents/Projects/TheRoad-I1/projects/ecedba66-2474-4dff-a1a4-7d13bac08e1d/episodes/5ea07552-480a-4fdd-a888-b6147ff239f5/images/background_chain/L04B02.png", "prompt": "Draw ONE storyboard layout frame, 16:9, filling the entire image edge-to-edge.\n\nSTYLE — LAYOUT-ONLY THUMBNAIL SKETCH: the lightest possible\ndirector's layout drawing. THIN, CLEAN, UNIFORM PENCIL CONTOUR\nLINES ONLY on pure white paper. Absolutely NO shading, NO tone, NO\nhatching, NO texture, NO grayscale fill, NO light-and-shadow, NO\nfacial features beyond a bare head outline, NO clothing detail, NO\nmaterial rendering. The frame contains ONLY: the camera framing,\neach figure as a simple correct-proportion human contour in its\nexact pose and screen position/size, and the few major set masses\nas plain outline shapes. It must read as a quick layout guide,\nnever as a finished drawing.\n\nThe attached photograph is the LOCATION where this shot happens.\nUse it ONLY as a LOOSE spatial reference — the rough set masses\nand where things sit relative to each other. You do NOT need to\nfollow it exactly: do not copy its camera, its lighting or its\ndetails. Choose the camera and framing the SHOT TEXT needs, and\ndraw only the few set masses that matter as plain outline shapes.\n\nTHE LOCATION: 옥탑방 내부 주거 공간\n\nSHOT TEXT (authoritative, Korean): 강민숙(민숙)의 손목 피부 위에 선명하게 자리 잡은 작은 원형 표식 클로즈업.\n\nCARRIED STATE (persist exactly): The clearly defined small circular mark remains fixed on Minsuk's wrist.\n\nABSOLUTELY NO TEXT anywhere: no captions, no numbers, no labels of any kind.", "skipped_reason": null, "error": null, "asset_id": "a958c7ec-b62f-410c-a246-dd4e68cd876c" }, "S7sh1": { "image_path": null, "plate_path": null, "prompt": null, "skipped_reason": "bgonly", "error": null }, "S7sh3": { "image_path": null, "plate_path": null, "prompt": null, "skipped_reason": "bgonly", "error": null }, "S7sh5": { "image_path": null, "plate_path": null, "prompt": null, "skipped_reason": "prev", "error": null }, "S8sh2": { "image_path": "/Users/manta/Documents/Projects/TheRoad-I1/projects/ecedba66-2474-4dff-a1a4-7d13bac08e1d/images/5ea07552-480a-4fdd-a888-b6147ff239f5/conti/conti_S8sh2.png", "plate_path": "/Users/manta/Documents/Projects/TheRoad-I1/projects/ecedba66-2474-4dff-a1a4-7d13bac08e1d/episodes/5ea07552-480a-4fdd-a888-b6147ff239f5/images/background_chain/L08B01.png", "prompt": "Draw ONE storyboard layout frame, 16:9, filling the entire image edge-to-edge.\n\nSTYLE — LAYOUT-ONLY THUMBNAIL SKETCH: the lightest possible\ndirector's layout drawing. THIN, CLEAN, UNIFORM PENCIL CONTOUR\nLINES ONLY on pure white paper. Absolutely NO shading, NO tone, NO\nhatching, NO texture, NO grayscale fill, NO light-and-shadow, NO\nfacial features beyond a bare head outline, NO clothing detail, NO\nmaterial rendering. The frame contains ONLY: the camera framing,\neach figure as a simple correct-proportion human contour in its\nexact pose and screen position/size, and the few major set masses\nas plain outline shapes. It must read as a quick layout guide,\nnever as a finished drawing.\n\nThe attached photograph is the LOCATION where this shot happens.\nUse it ONLY as a LOOSE spatial reference — the rough set masses\nand where things sit relative to each other. You do NOT need to\nfollow it exactly: do not copy its camera, its lighting or its\ndetails. Choose the camera and framing the SHOT TEXT needs, and\ndraw only the few set masses that matter as plain outline shapes.\n\nTHE LOCATION: 대형 마트 매장 내부\n\nSHOT TEXT (authoritative, Korean): 마트 안 상품 진열대 사이에서 미간을 살짝 찌푸리고 두리번거리는 각도로 얼굴을 돌린 수리영.\n\nCARRIED STATE (persist exactly): Suriyoung's bicycle remains parked outside the mart, and she continues to carry her smartphone.\n\nABSOLUTELY NO TEXT anywhere: no captions, no numbers, no labels of any kind.", "skipped_reason": null, "error": null, "asset_id": "f1642764-76e5-44cd-9800-35d0a444d158" }, "S8sh4": { "image_path": "/Users/manta/Documents/Projects/TheRoad-I1/projects/ecedba66-2474-4dff-a1a4-7d13bac08e1d/images/5ea07552-480a-4fdd-a888-b6147ff239f5/conti/conti_S8sh4.png", "plate_path": "/Users/manta/Documents/Projects/TheRoad-I1/projects/ecedba66-2474-4dff-a1a4-7d13bac08e1d/episodes/5ea07552-480a-4fdd-a888-b6147ff239f5/images/background_chain/L08B02.png", "prompt": "Draw ONE storyboard layout frame, 16:9, filling the entire image edge-to-edge.\n\nSTYLE — LAYOUT-ONLY THUMBNAIL SKETCH: the lightest possible\ndirector's layout drawing. THIN, CLEAN, UNIFORM PENCIL CONTOUR\nLINES ONLY on pure white paper. Absolutely NO shading, NO tone, NO\nhatching, NO texture, NO grayscale fill, NO light-and-shadow, NO\nfacial features beyond a bare head outline, NO clothing detail, NO\nmaterial rendering. The frame contains ONLY: the camera framing,\neach figure as a simple correct-proportion human contour in its\nexact pose and screen position/size, and the few major set masses\nas plain outline shapes. It must read as a quick layout guide,\nnever as a finished drawing.\n\nThe attached photograph is the LOCATION where this shot happens.\nUse it ONLY as a LOOSE spatial reference — the rough set masses\nand where things sit relative to each other. You do NOT need to\nfollow it exactly: do not copy its camera, its lighting or its\ndetails. Choose the camera and framing the SHOT TEXT needs, and\ndraw only the few set masses that matter as plain outline shapes.\n\nTHE LOCATION: 대형 마트 매장 내부\n\nSHOT TEXT (authoritative, Korean): 마트 밖 번화가 길가에 서서 스마트폰 화면을 빤히 내려다보는 수리영.\n\nCARRIED STATE (persist exactly): Her bicycle remains outside the mart while Suriyoung continues carrying and checking her smartphone.\n\nABSOLUTELY NO TEXT anywhere: no captions, no numbers, no labels of any kind.", "skipped_reason": null, "error": null, "asset_id": "80d1c6df-22fe-40f9-a561-91797a78b08d" }, "S12sh6": { "image_path": "/Users/manta/Documents/Projects/TheRoad-I1/projects/ecedba66-2474-4dff-a1a4-7d13bac08e1d/images/5ea07552-480a-4fdd-a888-b6147ff239f5/conti/conti_S12sh6.png", "plate_path": "/Users/manta/Documents/Projects/TheRoad-I1/projects/ecedba66-2474-4dff-a1a4-7d13bac08e1d/episodes/5ea07552-480a-4fdd-a888-b6147ff239f5/images/background_chain/L04B06.png", "prompt": "Draw ONE storyboard layout frame, 16:9, filling the entire image edge-to-edge.\n\nSTYLE — LAYOUT-ONLY THUMBNAIL SKETCH: the lightest possible\ndirector's layout drawing. THIN, CLEAN, UNIFORM PENCIL CONTOUR\nLINES ONLY on pure white paper. Absolutely NO shading, NO tone, NO\nhatching, NO texture, NO grayscale fill, NO light-and-shadow, NO\nfacial features beyond a bare head outline, NO clothing detail, NO\nmaterial rendering. The frame contains ONLY: the camera framing,\neach figure as a simple correct-proportion human contour in its\nexact pose and screen position/size, and the few major set masses\nas plain outline shapes. It must read as a quick layout guide,\nnever as a finished drawing.\n\nThe attached photograph is the LOCATION where this shot happens.\nUse it ONLY as a LOOSE spatial reference — the rough set masses\nand where things sit relative to each other. You do NOT need to\nfollow it exactly: do not copy its camera, its lighting or its\ndetails. Choose the camera and framing the SHOT TEXT needs, and\ndraw only the few set masses that matter as plain outline shapes.\n\nTHE LOCATION: 옥탑방 내부 주거 공간\n\nSHOT TEXT (authoritative, Korean): 살점이 덜렁거리고 거칠게 물어뜯긴 어깨 상처에서 피가 흐르는 강민숙(민숙)이 고개를 푹 떨구고 바닥에 앉아 있는 참혹한 모습.\n\nIMMOBILE CHARACTER POSE — CANONICAL (identical wherever this character appears in ANY panel; on any conflict THIS POSE WINS): Min-sook is seated motionless on the floor behind the curtain, with her torso slumped and her head hanging forward. One rigid hand grips a crumpled photograph, and her remaining limbs stay slack in the seated position.\n\nCARRIED STATE (persist exactly): Minsuk remains dead, seated motionless on the floor behind the curtain, with her torso slumped and her head hanging forward, her shoulder and collarbone torn open and bleeding; the room retains the bloodstains, pooled blood, and bloody footprints. Because she cannot grasp it later, one rigid hand already grips the crumpled photograph, and her wrist still bears the circular mark.\n\nABSOLUTELY NO TEXT anywhere: no captions, no numbers, no labels of any kind.", "skipped_reason": null, "error": null, "asset_id": "f4407137-0793-42a9-bca3-9f6ffdc35571" }, "S12sh9": { "image_path": null, "plate_path": null, "prompt": null, "skipped_reason": "prev", "error": null }, "S14sh2": { "image_path": null, "plate_path": null, "prompt": null, "skipped_reason": "bgonly", "error": null }, "S14sh3": { "image_path": null, "plate_path": null, "prompt": null, "skipped_reason": "prev", "error": null }, "S16sh1": { "image_path": "/Users/manta/Documents/Projects/TheRoad-I1/projects/ecedba66-2474-4dff-a1a4-7d13bac08e1d/images/5ea07552-480a-4fdd-a888-b6147ff239f5/conti/conti_S16sh1.png", "plate_path": "/Users/manta/Documents/Projects/TheRoad-I1/projects/ecedba66-2474-4dff-a1a4-7d13bac08e1d/episodes/5ea07552-480a-4fdd-a888-b6147ff239f5/images/background_chain/L11B02.png", "prompt": "Draw ONE storyboard layout frame, 16:9, filling the entire image edge-to-edge.\n\nSTYLE — LAYOUT-ONLY THUMBNAIL SKETCH: the lightest possible\ndirector's layout drawing. THIN, CLEAN, UNIFORM PENCIL CONTOUR\nLINES ONLY on pure white paper. Absolutely NO shading, NO tone, NO\nhatching, NO texture, NO grayscale fill, NO light-and-shadow, NO\nfacial features beyond a bare head outline, NO clothing detail, NO\nmaterial rendering. The frame contains ONLY: the camera framing,\neach figure as a simple correct-proportion human contour in its\nexact pose and screen position/size, and the few major set masses\nas plain outline shapes. It must read as a quick layout guide,\nnever as a finished drawing.\n\nThe attached photograph is the LOCATION where this shot happens.\nUse it ONLY as a LOOSE spatial reference — the rough set masses\nand where things sit relative to each other. You do NOT need to\nfollow it exactly: do not copy its camera, its lighting or its\ndetails. Choose the camera and framing the SHOT TEXT needs, and\ndraw only the few set masses that matter as plain outline shapes.\n\nTHE LOCATION: 경찰서 사무실\n\nSHOT TEXT (authoritative, Korean): 경찰서 사무실 안, 입술을 꽉 다문 채 스마트폰을 귀에 바짝 밀착시키고 선 혜수의 굳은 옆얼굴.\n\nCARRIED STATE (persist exactly): Hyesu continues carrying her smartphone and is still using Minsuk's saved number to call her without a response.\n\nABSOLUTELY NO TEXT anywhere: no captions, no numbers, no labels of any kind.", "skipped_reason": null, "error": null, "asset_id": "183dc533-ad36-4749-857e-1475710981db" }, "S18sh1": { "image_path": "/Users/manta/Documents/Projects/TheRoad-I1/projects/ecedba66-2474-4dff-a1a4-7d13bac08e1d/images/5ea07552-480a-4fdd-a888-b6147ff239f5/conti/conti_S18sh1.png", "plate_path": "/Users/manta/Documents/Projects/TheRoad-I1/projects/ecedba66-2474-4dff-a1a4-7d13bac08e1d/episodes/5ea07552-480a-4fdd-a888-b6147ff239f5/images/background_chain/L04B03.png", "prompt": "Draw ONE storyboard layout frame, 16:9, filling the entire image edge-to-edge.\n\nSTYLE — LAYOUT-ONLY THUMBNAIL SKETCH: the lightest possible\ndirector's layout drawing. THIN, CLEAN, UNIFORM PENCIL CONTOUR\nLINES ONLY on pure white paper. Absolutely NO shading, NO tone, NO\nhatching, NO texture, NO grayscale fill, NO light-and-shadow, NO\nfacial features beyond a bare head outline, NO clothing detail, NO\nmaterial rendering. The frame contains ONLY: the camera framing,\neach figure as a simple correct-proportion human contour in its\nexact pose and screen position/size, and the few major set masses\nas plain outline shapes. It must read as a quick layout guide,\nnever as a finished drawing.\n\nThe attached photograph is the LOCATION where this shot happens.\nUse it ONLY as a LOOSE spatial reference — the rough set masses\nand where things sit relative to each other. You do NOT need to\nfollow it exactly: do not copy its camera, its lighting or its\ndetails. Choose the camera and framing the SHOT TEXT needs, and\ndraw only the few set masses that matter as plain outline shapes.\n\nTHE LOCATION: 옥탑방 내부 주거 공간\n\nSHOT TEXT (authoritative, Korean): 밝은 낮, 안방 침대 모서리에 이마를 기댄 채 두 눈을 감고 앉아 있는 수리영의 상체.\n\nCARRIED STATE (persist exactly): The house remains vacant, and Suriyoung's room is still unnaturally clean with no corpse or blood present. Her doll-adorned backpack remains in the house until she takes it out later in the scene.\n\nABSOLUTELY NO TEXT anywhere: no captions, no numbers, no labels of any kind.", "skipped_reason": null, "error": null, "asset_id": "2e19b701-e551-48ef-958f-425d083d0622" }, "S18sh8": { "image_path": null, "plate_path": null, "prompt": null, "skipped_reason": "bgonly", "error": null }, "S18sh11": { "image_path": null, "plate_path": null, "prompt": null, "skipped_reason": "bgonly", "error": null }, "S19sh1": { "image_path": "/Users/manta/Documents/Projects/TheRoad-I1/projects/ecedba66-2474-4dff-a1a4-7d13bac08e1d/images/5ea07552-480a-4fdd-a888-b6147ff239f5/conti/conti_S19sh1.png", "plate_path": "/Users/manta/Documents/Projects/TheRoad-I1/projects/ecedba66-2474-4dff-a1a4-7d13bac08e1d/episodes/5ea07552-480a-4fdd-a888-b6147ff239f5/images/background_chain/L08B04.png", "prompt": "Draw ONE storyboard layout frame, 16:9, filling the entire image edge-to-edge.\n\nSTYLE — LAYOUT-ONLY THUMBNAIL SKETCH: the lightest possible\ndirector's layout drawing. THIN, CLEAN, UNIFORM PENCIL CONTOUR\nLINES ONLY on pure white paper. Absolutely NO shading, NO tone, NO\nhatching, NO texture, NO grayscale fill, NO light-and-shadow, NO\nfacial features beyond a bare head outline, NO clothing detail, NO\nmaterial rendering. The frame contains ONLY: the camera framing,\neach figure as a simple correct-proportion human contour in its\nexact pose and screen position/size, and the few major set masses\nas plain outline shapes. It must read as a quick layout guide,\nnever as a finished drawing.\n\nThe attached photograph is the LOCATION where this shot happens.\nUse it ONLY as a LOOSE spatial reference — the rough set masses\nand where things sit relative to each other. You do NOT need to\nfollow it exactly: do not copy its camera, its lighting or its\ndetails. Choose the camera and framing the SHOT TEXT needs, and\ndraw only the few set masses that matter as plain outline shapes.\n\nTHE LOCATION: 대형 마트 매장 내부\n\nSHOT TEXT (authoritative, Korean): 이른 아침 한산한 마트 안, 마트 남자 직원을 마주보고 꼿꼿이 서 있는 수리영의 정면.\n\nCARRIED STATE (persist exactly): Suriyoung's bicycle remains parked outside the mart, while her doll-adorned backpack and smartphone stay with her.\n\nABSOLUTELY NO TEXT anywhere: no captions, no numbers, no labels of any kind.", "skipped_reason": null, "error": null, "asset_id": "35f247df-3202-4950-b75c-f39f4f941a98" }, "S19sh4": { "image_path": "/Users/manta/Documents/Projects/TheRoad-I1/projects/ecedba66-2474-4dff-a1a4-7d13bac08e1d/images/5ea07552-480a-4fdd-a888-b6147ff239f5/conti/conti_S19sh4.png", "plate_path": "/Users/manta/Documents/Projects/TheRoad-I1/projects/ecedba66-2474-4dff-a1a4-7d13bac08e1d/episodes/5ea07552-480a-4fdd-a888-b6147ff239f5/images/background_chain/L08B03.png", "prompt": "Draw ONE storyboard layout frame, 16:9, filling the entire image edge-to-edge.\n\nSTYLE — LAYOUT-ONLY THUMBNAIL SKETCH: the lightest possible\ndirector's layout drawing. THIN, CLEAN, UNIFORM PENCIL CONTOUR\nLINES ONLY on pure white paper. Absolutely NO shading, NO tone, NO\nhatching, NO texture, NO grayscale fill, NO light-and-shadow, NO\nfacial features beyond a bare head outline, NO clothing detail, NO\nmaterial rendering. The frame contains ONLY: the camera framing,\neach figure as a simple correct-proportion human contour in its\nexact pose and screen position/size, and the few major set masses\nas plain outline shapes. It must read as a quick layout guide,\nnever as a finished drawing.\n\nThe attached photograph is the LOCATION where this shot happens.\nUse it ONLY as a LOOSE spatial reference — the rough set masses\nand where things sit relative to each other. You do NOT need to\nfollow it exactly: do not copy its camera, its lighting or its\ndetails. Choose the camera and framing the SHOT TEXT needs, and\ndraw only the few set masses that matter as plain outline shapes.\n\nTHE LOCATION: 대형 마트 매장 내부\n\nSHOT TEXT (authoritative, Korean): 마트 밖에서 스마트폰을 귀에 댄 채 시선을 허공에 둔 수리영의 옆얼굴.\n\nCARRIED STATE (persist exactly): Suriyoung continues carrying her backpack and smartphone while calling Hyesu; her bicycle remains parked at the mart.\n\nABSOLUTELY NO TEXT anywhere: no captions, no numbers, no labels of any kind.", "skipped_reason": null, "error": null, "asset_id": "77258fed-20e9-4e90-93c7-b1ea8dffb90c" }, "S20sh3": { "image_path": "/Users/manta/Documents/Projects/TheRoad-I1/projects/ecedba66-2474-4dff-a1a4-7d13bac08e1d/images/5ea07552-480a-4fdd-a888-b6147ff239f5/conti/conti_S20sh3.png", "plate_path": "/Users/manta/Documents/Projects/TheRoad-I1/projects/ecedba66-2474-4dff-a1a4-7d13bac08e1d/episodes/5ea07552-480a-4fdd-a888-b6147ff239f5/images/background_chain/L12B01.png", "prompt": "Draw ONE storyboard layout frame, 16:9, filling the entire image edge-to-edge.\n\nSTYLE — LAYOUT-ONLY THUMBNAIL SKETCH: the lightest possible\ndirector's layout drawing. THIN, CLEAN, UNIFORM PENCIL CONTOUR\nLINES ONLY on pure white paper. Absolutely NO shading, NO tone, NO\nhatching, NO texture, NO grayscale fill, NO light-and-shadow, NO\nfacial features beyond a bare head outline, NO clothing detail, NO\nmaterial rendering. The frame contains ONLY: the camera framing,\neach figure as a simple correct-proportion human contour in its\nexact pose and screen position/size, and the few major set masses\nas plain outline shapes. It must read as a quick layout guide,\nnever as a finished drawing.\n\nThe attached photograph is the LOCATION where this shot happens.\nUse it ONLY as a LOOSE spatial reference — the rough set masses\nand where things sit relative to each other. You do NOT need to\nfollow it exactly: do not copy its camera, its lighting or its\ndetails. Choose the camera and framing the SHOT TEXT needs, and\ndraw only the few set masses that matter as plain outline shapes.\n\nTHE LOCATION: 대형 마트 총무과\n\nSHOT TEXT (authoritative, Korean): 혜수와 나란히 서서 모니터를 노려보며 미간을 심하게 찌푸린 수리영의 불안한 얼굴.\n\nCARRIED STATE (persist exactly): Suriyoung still has her doll-adorned backpack and smartphone with her during the CCTV review.\n\nABSOLUTELY NO TEXT anywhere: no captions, no numbers, no labels of any kind.", "skipped_reason": null, "error": null, "asset_id": "4d54dd60-7fcc-44ac-bc41-7866117dc6df" }, "S21sh5": { "image_path": "/Users/manta/Documents/Projects/TheRoad-I1/projects/ecedba66-2474-4dff-a1a4-7d13bac08e1d/images/5ea07552-480a-4fdd-a888-b6147ff239f5/conti/conti_S21sh5.png", "plate_path": "/Users/manta/Documents/Projects/TheRoad-I1/projects/ecedba66-2474-4dff-a1a4-7d13bac08e1d/episodes/5ea07552-480a-4fdd-a888-b6147ff239f5/images/background_chain/L11B02.png", "prompt": "Draw ONE storyboard layout frame, 16:9, filling the entire image edge-to-edge.\n\nSTYLE — LAYOUT-ONLY THUMBNAIL SKETCH: the lightest possible\ndirector's layout drawing. THIN, CLEAN, UNIFORM PENCIL CONTOUR\nLINES ONLY on pure white paper. Absolutely NO shading, NO tone, NO\nhatching, NO texture, NO grayscale fill, NO light-and-shadow, NO\nfacial features beyond a bare head outline, NO clothing detail, NO\nmaterial rendering. The frame contains ONLY: the camera framing,\neach figure as a simple correct-proportion human contour in its\nexact pose and screen position/size, and the few major set masses\nas plain outline shapes. It must read as a quick layout guide,\nnever as a finished drawing.\n\nThe attached photograph is the LOCATION where this shot happens.\nUse it ONLY as a LOOSE spatial reference — the rough set masses\nand where things sit relative to each other. You do NOT need to\nfollow it exactly: do not copy its camera, its lighting or its\ndetails. Choose the camera and framing the SHOT TEXT needs, and\ndraw only the few set masses that matter as plain outline shapes.\n\nTHE LOCATION: 경찰서 사무실\n\nSHOT TEXT (authoritative, Korean): 책상 위 사진들을 향해 시선을 던진 채 두 눈을 부릅뜬 수리영의 굳은 표정.\n\nCARRIED STATE (persist exactly): The enlarged crime-scene photographs remain scattered across Hyesu's desk. Suriyoung continues carrying her backpack and smartphone.\n\nABSOLUTELY NO TEXT anywhere: no captions, no numbers, no labels of any kind.", "skipped_reason": null, "error": null, "asset_id": "f5860a06-24db-41c6-bce8-68b4d1bce875" }, "S21sh6": { "image_path": null, "plate_path": null, "prompt": null, "skipped_reason": "bgonly", "error": null }, "S21sh10": { "image_path": null, "plate_path": null, "prompt": null, "skipped_reason": "prev", "error": null }, "S23sh1": { "image_path": null, "plate_path": null, "prompt": null, "skipped_reason": "no_plate", "error": null }, "S23sh2": { "image_path": null, "plate_path": null, "prompt": null, "skipped_reason": "bgonly", "error": null }, "S25sh4": { "image_path": null, "plate_path": null, "prompt": null, "skipped_reason": "bgonly", "error": null }, "S25sh8": { "image_path": null, "plate_path": null, "prompt": null, "skipped_reason": "prev", "error": null }, "S27sh1": { "image_path": null, "plate_path": null, "prompt": null, "skipped_reason": "prev", "error": null }, "S29sh4": { "image_path": "/Users/manta/Documents/Projects/TheRoad-I1/projects/ecedba66-2474-4dff-a1a4-7d13bac08e1d/images/5ea07552-480a-4fdd-a888-b6147ff239f5/conti/conti_S29sh4.png", "plate_path": "/Users/manta/Documents/Projects/TheRoad-I1/projects/ecedba66-2474-4dff-a1a4-7d13bac08e1d/episodes/5ea07552-480a-4fdd-a888-b6147ff239f5/images/background_chain/L17B02.png", "prompt": "Draw ONE storyboard layout frame, 16:9, filling the entire image edge-to-edge.\n\nSTYLE — LAYOUT-ONLY THUMBNAIL SKETCH: the lightest possible\ndirector's layout drawing. THIN, CLEAN, UNIFORM PENCIL CONTOUR\nLINES ONLY on pure white paper. Absolutely NO shading, NO tone, NO\nhatching, NO texture, NO grayscale fill, NO light-and-shadow, NO\nfacial features beyond a bare head outline, NO clothing detail, NO\nmaterial rendering. The frame contains ONLY: the camera framing,\neach figure as a simple correct-proportion human contour in its\nexact pose and screen position/size, and the few major set masses\nas plain outline shapes. It must read as a quick layout guide,\nnever as a finished drawing.\n\nThe attached photograph is the LOCATION where this shot happens.\nUse it ONLY as a LOOSE spatial reference — the rough set masses\nand where things sit relative to each other. You do NOT need to\nfollow it exactly: do not copy its camera, its lighting or its\ndetails. Choose the camera and framing the SHOT TEXT needs, and\ndraw only the few set masses that matter as plain outline shapes.\n\nTHE LOCATION: 소형 어선 조타실\n\nSHOT TEXT (authoritative, Korean): 수리영의 검지손가락이 낡은 사진 속 여자아이 형상 표면 위에 맞닿아 있는 정적 구도.\n\nCARRIED STATE (persist exactly): The old photograph remains in Suriyoung's possession as she points to the child in it. The wheelhouse walls already bear the numerous circular marks that become visible moments later; her backpack, maps, note, and phone remain with her.\n\nABSOLUTELY NO TEXT anywhere: no captions, no numbers, no labels of any kind.", "skipped_reason": null, "error": null, "asset_id": "39ee58dc-9993-41e7-bad0-bb70cb714dbe" }, "S29sh5": { "image_path": "/Users/manta/Documents/Projects/TheRoad-I1/projects/ecedba66-2474-4dff-a1a4-7d13bac08e1d/images/5ea07552-480a-4fdd-a888-b6147ff239f5/conti/conti_S29sh5.png", "plate_path": "/Users/manta/Documents/Projects/TheRoad-I1/projects/ecedba66-2474-4dff-a1a4-7d13bac08e1d/episodes/5ea07552-480a-4fdd-a888-b6147ff239f5/images/background_chain/L17B01.png", "prompt": "Draw ONE storyboard layout frame, 16:9, filling the entire image edge-to-edge.\n\nSTYLE — LAYOUT-ONLY THUMBNAIL SKETCH: the lightest possible\ndirector's layout drawing. THIN, CLEAN, UNIFORM PENCIL CONTOUR\nLINES ONLY on pure white paper. Absolutely NO shading, NO tone, NO\nhatching, NO texture, NO grayscale fill, NO light-and-shadow, NO\nfacial features beyond a bare head outline, NO clothing detail, NO\nmaterial rendering. The frame contains ONLY: the camera framing,\neach figure as a simple correct-proportion human contour in its\nexact pose and screen position/size, and the few major set masses\nas plain outline shapes. It must read as a quick layout guide,\nnever as a finished drawing.\n\nThe attached photograph is the LOCATION where this shot happens.\nUse it ONLY as a LOOSE spatial reference — the rough set masses\nand where things sit relative to each other. You do NOT need to\nfollow it exactly: do not copy its camera, its lighting or its\ndetails. Choose the camera and framing the SHOT TEXT needs, and\ndraw only the few set masses that matter as plain outline shapes.\n\nTHE LOCATION: 소형 어선 조타실\n\nSHOT TEXT (authoritative, Korean): 두 눈동자가 크게 확장된 채 멍하게 굳어버린 창백한 인우의 얼굴 클로즈업.\n\nCARRIED STATE (persist exactly): The wheelhouse remains covered with numerous circular marks, including the red circle noticed by Suriyoung. The old photograph and Suriyoung's other belongings remain with her aboard the boat.\n\nABSOLUTELY NO TEXT anywhere: no captions, no numbers, no labels of any kind.", "skipped_reason": null, "error": null, "asset_id": "af0c10d3-91f4-4d94-b7e0-ab1fd21163c5" }, "S29sh8": { "image_path": "/Users/manta/Documents/Projects/TheRoad-I1/projects/ecedba66-2474-4dff-a1a4-7d13bac08e1d/images/5ea07552-480a-4fdd-a888-b6147ff239f5/conti/conti_S29sh8.png", "plate_path": "/Users/manta/Documents/Projects/TheRoad-I1/projects/ecedba66-2474-4dff-a1a4-7d13bac08e1d/episodes/5ea07552-480a-4fdd-a888-b6147ff239f5/images/background_chain/L17B01.png", "prompt": "Draw ONE storyboard layout frame, 16:9, filling the entire image edge-to-edge.\n\nSTYLE — LAYOUT-ONLY THUMBNAIL SKETCH: the lightest possible\ndirector's layout drawing. THIN, CLEAN, UNIFORM PENCIL CONTOUR\nLINES ONLY on pure white paper. Absolutely NO shading, NO tone, NO\nhatching, NO texture, NO grayscale fill, NO light-and-shadow, NO\nfacial features beyond a bare head outline, NO clothing detail, NO\nmaterial rendering. The frame contains ONLY: the camera framing,\neach figure as a simple correct-proportion human contour in its\nexact pose and screen position/size, and the few major set masses\nas plain outline shapes. It must read as a quick layout guide,\nnever as a finished drawing.\n\nThe attached photograph is the LOCATION where this shot happens.\nUse it ONLY as a LOOSE spatial reference — the rough set masses\nand where things sit relative to each other. You do NOT need to\nfollow it exactly: do not copy its camera, its lighting or its\ndetails. Choose the camera and framing the SHOT TEXT needs, and\ndraw only the few set masses that matter as plain outline shapes.\n\nTHE LOCATION: 소형 어선 조타실\n\nSHOT TEXT (authoritative, Korean): 눈이 반쯤 감긴 수리영의 고개가 꼿꼿이 선 인우의 어깨 위에 힘없이 얹혀 있는 찰나의 정지 상태.\n\nIMMOBILE CHARACTER POSE — CANONICAL (identical wherever this character appears in ANY panel; on any conflict THIS POSE WINS): Su-ri-young’s limp body slumps toward the upright In-woo, with her half-closed-eyed head resting heavily on his shoulder. Her limbs are unbraced as his shoulder supports her collapsing upper body.\n\nCARRIED STATE (persist exactly): Su-ri-young is now unconscious. Su-ri-young’s limp body slumps toward the upright In-woo, with her half-closed-eyed head resting heavily on his shoulder. Her limbs are unbraced as his shoulder supports her collapsing upper body; the old photograph, her backpack and maps, the coordinate note, and her smartphone remain with her aboard the boat. The wheelhouse's numerous circular marks remain unchanged.\n\nABSOLUTELY NO TEXT anywhere: no captions, no numbers, no labels of any kind.", "skipped_reason": null, "error": null, "asset_id": "abb4cdb2-bb89-4950-9a3d-387577ff1e72" } }, "map_conti": {}, "lane_conti": { "S1sh3": { "lane": "map_marker", "group_id": "bg_mountain_forest", "segment_id": "SEG_01", "status": "ok", "image_path": "/Users/manta/Documents/Projects/TheRoad-I1/projects/ecedba66-2474-4dff-a1a4-7d13bac08e1d/images/5ea07552-480a-4fdd-a888-b6147ff239f5/conti/lane_sketch_S1sh3.png", "control_path": "/Users/manta/Documents/Projects/TheRoad-I1/projects/ecedba66-2474-4dff-a1a4-7d13bac08e1d/images/5ea07552-480a-4fdd-a888-b6147ff239f5/conti/lane_control_S1sh3.png", "base_map_path": "/Users/manta/Documents/Projects/TheRoad-I1/projects/ecedba66-2474-4dff-a1a4-7d13bac08e1d/images/background_chain/canon_bg_mountain_forest_map.png", "base_map_asset_id": "6c0aad1a-afa9-4f53-9325-7c7c34c36280", "seed_path": null, "seed_asset_id": null, "geometry": { "camera": { "look_target": { "x": 0.535, "y": 0.573 }, "origin": { "x": 0.553, "y": 0.596 } }, "entity_placements": [ { "slot": "E1", "subject_en": "terrified fleeing man", "x": 0.535, "y": 0.573 } ], "rationale_ko": "남성을 A3의 수풀 사이 숲 바닥에 배치했다. 남성이 화면 아래쪽으로 도주하다 어두운 뒤편인 북쪽을 돌아보는 순간으로 설정하고, 카메라는 진행 방향 기준 앞왼쪽에 해당하는 남동쪽에서 눈높이의 근접 촬영 축을 잡도록 배치했다." }, "leakage": { "has_marker_leakage": false, "details_ko": "마커 기호나 평면도 유출이 없습니다." }, "prompt": "Convert the attached TOP-DOWN MARKER MAP into a single CAMERA-VIEW\nstoryboard frame (16:9), seen from the camera marked on the map.\n\nJurisdiction of sources — follow both, they do not conflict:\n- The map is the SOT for XY POSITIONS and horizontal heading ONLY.\n Each red circle labeled E1, E2, ... is where a subject stands; the\n blue dot labeled CAM is the camera position; the blue wedge is its\n horizontal viewing direction and field of view. Subjects outside\n the wedge are out of frame. Distances, directions and relative\n depths between subjects come from the map.\n- The CAMERA DIRECTION text below (when present) is the SOT for the\n VERTICAL angle, elevation, tilt and framing (eye-level, high angle,\n overhead crane, low angle, ...). If it asks for a high or overhead\n angle, draw that angle — the map being top-down does NOT mean the\n frame must be eye-level.\n\nSTYLE — LIGHT STORYBOARD FRAME: thin uniform dark outlines on plain\npaper. No shading, no hatching, no texture, no facial features beyond\na minimal head shape, no costume detail. Human figures are correctly\nproportioned full-body outlines in their canonical poses. Environment\nis reduced to a few placement lines (ground line, key landmark\nsilhouettes) — just enough to read positions and depth.\n\nMAP SLOT MEANINGS:\nE1 = terrified fleeing man\n\nCAMERA DIRECTION (vertical angle / elevation / framing SOT): At the fleeing man's eye level and slightly off his front-left side, the pan settles into a face-dominant close frame as his head snaps toward the darkness behind him. His terrified face occupies the left two-thirds, while a narrow strip of the area beyond his turned head remains visible at frame right, preserving the axis for the following pivot.\n\nTHE LOCATION: 해 질 무렵의 우거진 숲속\n\nSHOT TEXT (authoritative, Korean): 눈을 부릅뜨고 입을 벌린 채, 어두운 뒤편을 향해 고개를 돌린 한국인 남성의 겁에 질린 얼굴.\n\nCRITICAL — THE MAP IS A STAGING DIAGRAM, NOT CONTENT: do NOT draw the\nmap itself, and do NOT draw any circles, wedges, dots, slot codes\n(E1, E2, CAM) or any other marker symbols in the frame. The output is\na clean storyboard drawing of the scene as seen by the camera only.\n\nNo text, no letters, no numbers, no captions anywhere in the image.", "reused": true, "control_asset_id": "e17f2e93-1492-4942-afd6-3fce749b621d", "asset_id": "83479135-e04a-45b1-b3e8-113803fa256f" }, "S1sh5": { "lane": "map_marker", "group_id": "bg_mountain_forest", "segment_id": "SEG_01", "status": "ok", "image_path": "/Users/manta/Documents/Projects/TheRoad-I1/projects/ecedba66-2474-4dff-a1a4-7d13bac08e1d/images/5ea07552-480a-4fdd-a888-b6147ff239f5/conti/lane_sketch_S1sh5.png", "control_path": "/Users/manta/Documents/Projects/TheRoad-I1/projects/ecedba66-2474-4dff-a1a4-7d13bac08e1d/images/5ea07552-480a-4fdd-a888-b6147ff239f5/conti/lane_control_S1sh5.png", "base_map_path": "/Users/manta/Documents/Projects/TheRoad-I1/projects/ecedba66-2474-4dff-a1a4-7d13bac08e1d/images/background_chain/canon_bg_mountain_forest_map.png", "base_map_asset_id": "6c0aad1a-afa9-4f53-9325-7c7c34c36280", "seed_path": null, "seed_asset_id": null, "geometry": { "camera": { "look_target": { "x": 0.538, "y": 0.568 }, "origin": { "x": 0.492, "y": 0.625 } }, "entity_placements": [ { "slot": "E1", "subject_en": "black-clad human figure", "x": 0.537, "y": 0.568 }, { "slot": "E2", "subject_en": "fallen Korean man", "x": 0.532, "y": 0.583 } ], "rationale_ko": "A3의 수풀 사이 숲 바닥에 남성을 쓰러진 상태로 두고, 검은 인영을 남성의 상체 바로 위에 겹치듯 배치해 체중으로 짓누르는 순간을 표현했다. 카메라는 두 인물의 남서쪽 가까운 지면 높이에 두어 남성이 화면 하단, 검은 인영이 상단 중앙에 읽히는 비스듬한 시선을 형성했다." }, "leakage": { "has_marker_leakage": false, "details_ko": "이미지에 지도 마커나 코드가 포함되어 있지 않습니다." }, "prompt": "Convert the attached TOP-DOWN MARKER MAP into a single CAMERA-VIEW\nstoryboard frame (16:9), seen from the camera marked on the map.\n\nJurisdiction of sources — follow both, they do not conflict:\n- The map is the SOT for XY POSITIONS and horizontal heading ONLY.\n Each red circle labeled E1, E2, ... is where a subject stands; the\n blue dot labeled CAM is the camera position; the blue wedge is its\n horizontal viewing direction and field of view. Subjects outside\n the wedge are out of frame. Distances, directions and relative\n depths between subjects come from the map.\n- The CAMERA DIRECTION text below (when present) is the SOT for the\n VERTICAL angle, elevation, tilt and framing (eye-level, high angle,\n overhead crane, low angle, ...). If it asks for a high or overhead\n angle, draw that angle — the map being top-down does NOT mean the\n frame must be eye-level.\n\nSTYLE — LIGHT STORYBOARD FRAME: thin uniform dark outlines on plain\npaper. No shading, no hatching, no texture, no facial features beyond\na minimal head shape, no costume detail. Human figures are correctly\nproportioned full-body outlines in their canonical poses. Environment\nis reduced to a few placement lines (ground line, key landmark\nsilhouettes) — just enough to read positions and depth.\n\nMAP SLOT MEANINGS:\nE1 = black-clad human figure\nE2 = fallen Korean man\n\nCAMERA DIRECTION (vertical angle / elevation / framing SOT): The handheld camera settles near ground level at an oblique angle toward the fallen man's upper body, retaining the unstable residue of the impact without leaving the flow's lowest point. The pinned man lies across the lower half of the frame while the black figure bears down from the upper center, making its downward force and his restrained torso readable at once.\n\nTHE LOCATION: 해 질 무렵의 우거진 숲속\n\nSHOT TEXT (authoritative, Korean): 바닥에 넘어진 한국인 남성 위로 검은 인영이 체중을 실어 거칠게 짓누르고, 남자의 상체가 바닥에 억눌려 있는 순간.\n\nCRITICAL — THE MAP IS A STAGING DIAGRAM, NOT CONTENT: do NOT draw the\nmap itself, and do NOT draw any circles, wedges, dots, slot codes\n(E1, E2, CAM) or any other marker symbols in the frame. The output is\na clean storyboard drawing of the scene as seen by the camera only.\n\nNo text, no letters, no numbers, no captions anywhere in the image.", "reused": true, "control_asset_id": "7180b3c6-1ceb-433e-9606-549014445c58", "asset_id": "7b4cabbc-759a-43f2-9314-6297b50d481b" }, "S2sh1": { "lane": "map_marker", "group_id": "bg_urban_alley", "segment_id": "SEG_01", "status": "ok", "image_path": "/Users/manta/Documents/Projects/TheRoad-I1/projects/ecedba66-2474-4dff-a1a4-7d13bac08e1d/images/5ea07552-480a-4fdd-a888-b6147ff239f5/conti/lane_sketch_S2sh1.png", "control_path": "/Users/manta/Documents/Projects/TheRoad-I1/projects/ecedba66-2474-4dff-a1a4-7d13bac08e1d/images/5ea07552-480a-4fdd-a888-b6147ff239f5/conti/lane_control_S2sh1.png", "base_map_path": "/Users/manta/Documents/Projects/TheRoad-I1/projects/ecedba66-2474-4dff-a1a4-7d13bac08e1d/images/background_chain/canon_bg_urban_alley_map.png", "base_map_asset_id": "4f08b7ec-dc04-4e47-96e1-95bc24085f9f", "seed_path": null, "seed_asset_id": null, "geometry": { "camera": { "look_target": { "x": 0.452, "y": 0.604 }, "origin": { "x": 0.548, "y": 0.782 } }, "entity_placements": [ { "slot": "E1", "subject_en": "woman leaning against utility pole", "x": 0.451, "y": 0.602 }, { "slot": "E2", "subject_en": "man pressed closely against woman", "x": 0.469, "y": 0.635 } ], "rationale_ko": "여성은 골목 좌측의 B1 전봇대 바로 옆에 등을 댄 위치로 두고, 남성은 골목 중앙 쪽에서 여성과 거의 겹치도록 밀착 배치했다. 카메라는 골목 아래쪽에서 수 미터 떨어진 채 전봇대 방향을 대각선 정면으로 바라보며, 남성이 여성 앞에 중첩되어 보이도록 시선을 맞췄다." }, "leakage": { "has_marker_leakage": false, "details_ko": "불필요한 마커 기호나 평면도 형태가 없습니다." }, "prompt": "Convert the attached TOP-DOWN MARKER MAP into a single CAMERA-VIEW\nstoryboard frame (16:9), seen from the camera marked on the map.\n\nJurisdiction of sources — follow both, they do not conflict:\n- The map is the SOT for XY POSITIONS and horizontal heading ONLY.\n Each red circle labeled E1, E2, ... is where a subject stands; the\n blue dot labeled CAM is the camera position; the blue wedge is its\n horizontal viewing direction and field of view. Subjects outside\n the wedge are out of frame. Distances, directions and relative\n depths between subjects come from the map.\n- The CAMERA DIRECTION text below (when present) is the SOT for the\n VERTICAL angle, elevation, tilt and framing (eye-level, high angle,\n overhead crane, low angle, ...). If it asks for a high or overhead\n angle, draw that angle — the map being top-down does NOT mean the\n frame must be eye-level.\n\nSTYLE — LIGHT STORYBOARD FRAME: thin uniform dark outlines on plain\npaper. No shading, no hatching, no texture, no facial features beyond\na minimal head shape, no costume detail. Human figures are correctly\nproportioned full-body outlines in their canonical poses. Environment\nis reduced to a few placement lines (ground line, key landmark\nsilhouettes) — just enough to read positions and depth.\n\nMAP SLOT MEANINGS:\nE1 = woman leaning against utility pole\nE2 = man pressed closely against woman\n\nCAMERA DIRECTION (vertical angle / elevation / framing SOT): From chest height several meters away, the camera begins its dolly toward the utility pole on a frontal diagonal, holding both figures within the alley layout. The woman is compressed against the pole in the left midground and the man overlaps her from center frame, with neither body isolated from the other.\n\nTHE LOCATION: 도시 뒷골목 범죄 현장\n\nSHOT TEXT (authoritative, Korean): 희미한 가로등 불빛 아래, 전봇대에 등을 기댄 한국인 여성과 그 위에 겹쳐 서 있는 한국인 남성의 밀착된 상태.\n\nCRITICAL — THE MAP IS A STAGING DIAGRAM, NOT CONTENT: do NOT draw the\nmap itself, and do NOT draw any circles, wedges, dots, slot codes\n(E1, E2, CAM) or any other marker symbols in the frame. The output is\na clean storyboard drawing of the scene as seen by the camera only.\n\nNo text, no letters, no numbers, no captions anywhere in the image.", "reused": true, "control_asset_id": "90013481-8bad-4afe-ad73-5a22747bea8f", "asset_id": "06f6a067-16fc-4d3a-973f-212b4d9baa5e" }, "S2sh4": { "lane": "map_marker", "group_id": "bg_urban_alley", "segment_id": "SEG_01", "status": "ok", "image_path": "/Users/manta/Documents/Projects/TheRoad-I1/projects/ecedba66-2474-4dff-a1a4-7d13bac08e1d/images/5ea07552-480a-4fdd-a888-b6147ff239f5/conti/lane_sketch_S2sh4.png", "control_path": "/Users/manta/Documents/Projects/TheRoad-I1/projects/ecedba66-2474-4dff-a1a4-7d13bac08e1d/images/5ea07552-480a-4fdd-a888-b6147ff239f5/conti/lane_control_S2sh4.png", "base_map_path": "/Users/manta/Documents/Projects/TheRoad-I1/projects/ecedba66-2474-4dff-a1a4-7d13bac08e1d/images/background_chain/canon_bg_urban_alley_map.png", "base_map_asset_id": "4f08b7ec-dc04-4e47-96e1-95bc24085f9f", "seed_path": null, "seed_asset_id": null, "geometry": { "camera": { "look_target": { "x": 0.477, "y": 0.607 }, "origin": { "x": 0.482, "y": 0.63 } }, "entity_placements": [ { "slot": "E1", "subject_en": "collapsed woman", "x": 0.477, "y": 0.607 } ], "rationale_ko": "여성을 골목 통로(A1)의 아스팔트 위이자 사고 지점 전봇대(B1) 바로 옆에 배치했다. 카메라는 여성의 상체 가까이에서 머리와 몸통을 중심으로 급경사 하향 촬영하는 위치로 설정해 주변 아스팔트도 프레임에 남도록 했다." }, "leakage": { "has_marker_leakage": false, "details_ko": "마커 기호나 텍스트가 발견되지 않았습니다." }, "prompt": "Convert the attached TOP-DOWN MARKER MAP into a single CAMERA-VIEW\nstoryboard frame (16:9), seen from the camera marked on the map.\n\nJurisdiction of sources — follow both, they do not conflict:\n- The map is the SOT for XY POSITIONS and horizontal heading ONLY.\n Each red circle labeled E1, E2, ... is where a subject stands; the\n blue dot labeled CAM is the camera position; the blue wedge is its\n horizontal viewing direction and field of view. Subjects outside\n the wedge are out of frame. Distances, directions and relative\n depths between subjects come from the map.\n- The CAMERA DIRECTION text below (when present) is the SOT for the\n VERTICAL angle, elevation, tilt and framing (eye-level, high angle,\n overhead crane, low angle, ...). If it asks for a high or overhead\n angle, draw that angle — the map being top-down does NOT mean the\n frame must be eye-level.\n\nSTYLE — LIGHT STORYBOARD FRAME: thin uniform dark outlines on plain\npaper. No shading, no hatching, no texture, no facial features beyond\na minimal head shape, no costume detail. Human figures are correctly\nproportioned full-body outlines in their canonical poses. Environment\nis reduced to a few placement lines (ground line, key landmark\nsilhouettes) — just enough to read positions and depth.\n\nMAP SLOT MEANINGS:\nE1 = collapsed woman\n\nCAMERA DIRECTION (vertical angle / elevation / framing SOT): The descending camera completes its move less than a meter above the asphalt and holds a steep downward angle over the woman's slack upper body. Her bent head and torso occupy the central foreground, with enough asphalt retained around her to register the finality of the collapse.\n\nTHE LOCATION: 도시 뒷골목 범죄 현장\n\nSHOT TEXT (authoritative, Korean): 아스팔트 바닥에 고개가 꺾인 채 늘어져 있는 여성의 상체.\n\nCRITICAL — THE MAP IS A STAGING DIAGRAM, NOT CONTENT: do NOT draw the\nmap itself, and do NOT draw any circles, wedges, dots, slot codes\n(E1, E2, CAM) or any other marker symbols in the frame. The output is\na clean storyboard drawing of the scene as seen by the camera only.\n\nNo text, no letters, no numbers, no captions anywhere in the image.", "reused": true, "control_asset_id": "428fc875-46f1-426c-b2e6-cbb673a6e721", "asset_id": "6fd3947a-b017-4cf7-abe2-1301131508f7" }, "S3sh2": { "lane": "map_marker", "group_id": "bg_urban_alley", "segment_id": "SEG_02", "status": "ok", "image_path": "/Users/manta/Documents/Projects/TheRoad-I1/projects/ecedba66-2474-4dff-a1a4-7d13bac08e1d/images/5ea07552-480a-4fdd-a888-b6147ff239f5/conti/lane_sketch_S3sh2.png", "control_path": "/Users/manta/Documents/Projects/TheRoad-I1/projects/ecedba66-2474-4dff-a1a4-7d13bac08e1d/images/5ea07552-480a-4fdd-a888-b6147ff239f5/conti/lane_control_S3sh2.png", "base_map_path": "/Users/manta/Documents/Projects/TheRoad-I1/projects/ecedba66-2474-4dff-a1a4-7d13bac08e1d/images/background_chain/canon_bg_urban_alley_map.png", "base_map_asset_id": "4f08b7ec-dc04-4e47-96e1-95bc24085f9f", "seed_path": null, "seed_asset_id": null, "geometry": { "camera": { "look_target": { "x": 0.52, "y": 0.57 }, "origin": { "x": 0.52, "y": 0.72 } }, "entity_placements": [ { "slot": "E1", "subject_en": "standing female lead detective", "x": 0.52, "y": 0.57 }, { "slot": "E2", "subject_en": "standing detective", "x": 0.47, "y": 0.55 }, { "slot": "E3", "subject_en": "standing detective", "x": 0.58, "y": 0.55 }, { "slot": "E4", "subject_en": "standing detective", "x": 0.46, "y": 0.59 }, { "slot": "E5", "subject_en": "standing detective", "x": 0.59, "y": 0.59 }, { "slot": "E6", "subject_en": "covered body", "x": 0.49, "y": 0.64 } ], "rationale_ko": "A1 골목 통로의 B1 사건 현장용 전신주 인접부에 지휘 형사를 중심으로 세우고, 형사들을 좌우 가장자리에 분산했다. 시신은 그녀의 전방 아래쪽에 비스듬히 놓아 가리키는 팔이 화면 하단으로 향하게 했다. 카메라는 골목 남쪽에서 북쪽을 바라보며 지휘 형사의 정면과 형사 무리를 눈높이로 포착한다." }, "leakage": { "has_marker_leakage": false, "details_ko": "스케치에 원형 마커, 쐐기 모양, 코드 기호 등이 전혀 포함되지 않았습니다." }, "prompt": "Convert the attached TOP-DOWN MARKER MAP into a single CAMERA-VIEW\nstoryboard frame (16:9), seen from the camera marked on the map.\n\nJurisdiction of sources — follow both, they do not conflict:\n- The map is the SOT for XY POSITIONS and horizontal heading ONLY.\n Each red circle labeled E1, E2, ... is where a subject stands; the\n blue dot labeled CAM is the camera position; the blue wedge is its\n horizontal viewing direction and field of view. Subjects outside\n the wedge are out of frame. Distances, directions and relative\n depths between subjects come from the map.\n- The CAMERA DIRECTION text below (when present) is the SOT for the\n VERTICAL angle, elevation, tilt and framing (eye-level, high angle,\n overhead crane, low angle, ...). If it asks for a high or overhead\n angle, draw that angle — the map being top-down does NOT mean the\n frame must be eye-level.\n\nSTYLE — LIGHT STORYBOARD FRAME: thin uniform dark outlines on plain\npaper. No shading, no hatching, no texture, no facial features beyond\na minimal head shape, no costume detail. Human figures are correctly\nproportioned full-body outlines in their canonical poses. Environment\nis reduced to a few placement lines (ground line, key landmark\nsilhouettes) — just enough to read positions and depth.\n\nMAP SLOT MEANINGS:\nE1 = standing female lead detective\nE2 = standing detective\nE3 = standing detective\nE4 = standing detective\nE5 = standing detective\nE6 = covered body\n\nCAMERA DIRECTION (vertical angle / elevation / framing SOT): Presented as daytime regional-news footage, the completed lateral track arrives at standing eye level directly in front of 혜수, with mild broadcast-image compression framing her at the center of the detective group. Her extended pointing arm cuts diagonally toward the lower edge where the body lies, while surrounding detectives remain secondary along both margins.\n\nTHE LOCATION: 도시 뒷골목 범죄 현장\n\nSHOT TEXT (authoritative, Korean): 형사 무리 한가운데 꼿꼿이 서서 손가락으로 바닥의 시신 쪽을 가리키고 있는 혜수의 정면.\n\nCRITICAL — THE MAP IS A STAGING DIAGRAM, NOT CONTENT: do NOT draw the\nmap itself, and do NOT draw any circles, wedges, dots, slot codes\n(E1, E2, CAM) or any other marker symbols in the frame. The output is\na clean storyboard drawing of the scene as seen by the camera only.\n\nNo text, no letters, no numbers, no captions anywhere in the image.", "reused": true, "control_asset_id": "1c4e4e06-83fe-415f-878e-f02623c89bb3", "asset_id": "5f3b55d8-2467-4aad-9bc2-b039a917b44f" }, "S4sh1": { "lane": "structure_plate", "group_id": "bg_rooftop_residence", "segment_id": "SEG_01", "sketch_attempts": 1, "status": "ok", "image_path": "/Users/manta/Documents/Projects/TheRoad-I1/projects/ecedba66-2474-4dff-a1a4-7d13bac08e1d/images/5ea07552-480a-4fdd-a888-b6147ff239f5/conti/lane_sketch_S4sh1.png", "control_path": "/Users/manta/Documents/Projects/TheRoad-I1/projects/ecedba66-2474-4dff-a1a4-7d13bac08e1d/images/5ea07552-480a-4fdd-a888-b6147ff239f5/conti/lane_control_S4sh1.png", "base_map_path": "/Users/manta/Documents/Projects/TheRoad-I1/projects/ecedba66-2474-4dff-a1a4-7d13bac08e1d/images/background_chain/siteplan_bg_rooftop_residence.png", "base_map_asset_id": "08a2ad21-a63a-48bd-a8d8-ee6e1cb15320", "seed_path": "/Users/manta/Documents/Projects/TheRoad-I1/projects/ecedba66-2474-4dff-a1a4-7d13bac08e1d/images/background_chain/seed_bg_rooftop_residence_sel.png", "seed_asset_id": "2910d596-ddb4-4946-b782-4d6cc0e2a768", "geometry": { "camera": { "look_target": { "x": 0.642, "y": 0.398 }, "origin": { "x": 0.642, "y": 0.787 } }, "entity_placements": [ { "slot": "E1", "subject_en": "rooftop room roof", "x": 0.642, "y": 0.398 } ], "rationale_ko": "옥탑방 외부 구조(D2)의 지붕 중심을 화면 중심으로 삼고, 카메라는 후정 상공 쪽에서 옥탑방을 향하도록 배치했다. 높은 고도의 넓은 부감으로 옥탑방 지붕이 중앙 3분의 1보다 작게 보이며, 주변 옥상 데크(D1), 물탱크(D5), 경계벽(D6), 빌라 일부가 함께 드러나는 구도다." }, "geometry_attempts": 1, "leakage": { "has_marker_leakage": false, "details_ko": "도면 기호나 문자 없이 건물의 입체적인 모습이 잘 표현되었습니다." }, "prompt": "Convert the attached TOP-DOWN MARKER MAP into a single CAMERA-VIEW\nstoryboard frame (16:9), seen from the camera marked on the map.\n\nJurisdiction of sources — follow both, they do not conflict:\n- The map is the SOT for XY POSITIONS and horizontal heading ONLY.\n Each red circle labeled E1, E2, ... is where a subject stands; the\n blue dot labeled CAM is the camera position; the blue wedge is its\n horizontal viewing direction and field of view. Subjects outside\n the wedge are out of frame. Distances, directions and relative\n depths between subjects come from the map.\n- The CAMERA DIRECTION text below (when present) is the SOT for the\n VERTICAL angle, elevation, tilt and framing (eye-level, high angle,\n overhead crane, low angle, ...). If it asks for a high or overhead\n angle, draw that angle — the map being top-down does NOT mean the\n frame must be eye-level.\n\nSTYLE — LIGHT STORYBOARD FRAME: thin uniform dark outlines on plain\npaper. No shading, no hatching, no texture, no facial features beyond\na minimal head shape, no costume detail. Human figures are correctly\nproportioned full-body outlines in their canonical poses. Environment\nis reduced to a few placement lines (ground line, key landmark\nsilhouettes) — just enough to read positions and depth.\n\nMAP SLOT MEANINGS:\nE1 = rooftop room roof\n\nThe SECOND attached image is the confirmed photograph of the fixed\nstructure on this site. It is the STRUCTURE LOOK authority: wherever\nthe structure appears in the sketch, its shape, proportions, openings\nand silhouette must match this photograph (rendered only as the same\nsimple line-frame style as everything else). The FIRST attached\nstaging map remains the sole authority for positions, camera placement\nand heading.\n\nCAMERA DIRECTION (vertical angle / elevation / framing SOT): The continuous aerial approach ends in a high overhead wide frame centered on the rooftop-room roof. The roof occupies less than the central third, while the surrounding rooftop and neighboring portions of the old villa remain visible as scale and location references.\n\nTHE LOCATION: 다세대 빌라 옥상 및 옥탑방 외부\n\nSHOT TEXT (authoritative, Korean): 낮의 맑은 하늘 아래, 오래된 변두리 다세대 빌라 옥상 위의 옥탑방 지붕이 내려다보이는 넓은 부감 쇼트.\n\nCRITICAL — THE MAP IS A STAGING DIAGRAM, NOT CONTENT: do NOT draw the\nmap itself, and do NOT draw any circles, wedges, dots, slot codes\n(E1, E2, CAM) or any other marker symbols in the frame. The output is\na clean storyboard drawing of the scene as seen by the camera only.\n\nNo text, no letters, no numbers, no captions anywhere in the image.", "control_asset_id": "01733229-f5a6-4e01-92bd-b61d1e78a94f", "asset_id": "9f1c6b89-9303-42ff-8d86-a145025a6a39" }, "S6sh1": { "lane": "map_marker", "group_id": "bg_coastal_road", "segment_id": "SEG_01", "status": "ok", "image_path": "/Users/manta/Documents/Projects/TheRoad-I1/projects/ecedba66-2474-4dff-a1a4-7d13bac08e1d/images/5ea07552-480a-4fdd-a888-b6147ff239f5/conti/lane_sketch_S6sh1.png", "control_path": "/Users/manta/Documents/Projects/TheRoad-I1/projects/ecedba66-2474-4dff-a1a4-7d13bac08e1d/images/5ea07552-480a-4fdd-a888-b6147ff239f5/conti/lane_control_S6sh1.png", "base_map_path": "/Users/manta/Documents/Projects/TheRoad-I1/projects/ecedba66-2474-4dff-a1a4-7d13bac08e1d/images/background_chain/canon_bg_coastal_road_map.png", "base_map_asset_id": "a27f22eb-882a-4c81-902e-837555b3df77", "seed_path": null, "seed_asset_id": null, "geometry": { "camera": { "look_target": { "x": 0.56, "y": 0.59 }, "origin": { "x": 0.63, "y": 0.55 } }, "entity_placements": [ { "slot": "E1", "subject_en": "young woman riding a bicycle", "x": 0.56, "y": 0.59 } ], "rationale_ko": "자전거를 탄 인물을 A1 해안 도로의 중앙부에 두고 북쪽으로 주행하는 순간으로 배치했다. 카메라는 같은 도로 위에서 진행 방향 기준 옆쪽이자 약간 앞에 위치해 가슴 높이로 병행 추적하며, 페달을 밟는 동작과 뒤로 날리는 머리카락·옷깃이 드러나는 3/4 와이드 구도를 형성한다." }, "leakage": { "has_marker_leakage": false, "details_ko": "스케치에 마커 기호나 텍스트 누출이 없습니다." }, "prompt": "Convert the attached TOP-DOWN MARKER MAP into a single CAMERA-VIEW\nstoryboard frame (16:9), seen from the camera marked on the map.\n\nJurisdiction of sources — follow both, they do not conflict:\n- The map is the SOT for XY POSITIONS and horizontal heading ONLY.\n Each red circle labeled E1, E2, ... is where a subject stands; the\n blue dot labeled CAM is the camera position; the blue wedge is its\n horizontal viewing direction and field of view. Subjects outside\n the wedge are out of frame. Distances, directions and relative\n depths between subjects come from the map.\n- The CAMERA DIRECTION text below (when present) is the SOT for the\n VERTICAL angle, elevation, tilt and framing (eye-level, high angle,\n overhead crane, low angle, ...). If it asks for a high or overhead\n angle, draw that angle — the map being top-down does NOT mean the\n frame must be eye-level.\n\nSTYLE — LIGHT STORYBOARD FRAME: thin uniform dark outlines on plain\npaper. No shading, no hatching, no texture, no facial features beyond\na minimal head shape, no costume detail. Human figures are correctly\nproportioned full-body outlines in their canonical poses. Environment\nis reduced to a few placement lines (ground line, key landmark\nsilhouettes) — just enough to read positions and depth.\n\nMAP SLOT MEANINGS:\nE1 = young woman riding a bicycle\n\nCAMERA DIRECTION (vertical angle / elevation / framing SOT): Tracking at chest height beside and slightly ahead of the bicycle, the camera holds 수리영 in a dynamic three-quarter wide frame as her foot presses the pedal downward. She occupies the middle-left while the bicycle and road extend toward frame right; her raised chin, trailing hair, and moving collar preserve the forward momentum without breaking the parallel camera path.\n\nTHE LOCATION: 해안가 도로\n\nSHOT TEXT (authoritative, Korean): 이어폰을 낀 수리영이 해안가 도로에서 자전거 페달을 밟아 내리는 중간 찰나, 머리카락과 옷깃이 뒤로 휘날리며 고개를 살짝 위로 들고 있는 상태.\n\nCRITICAL — THE MAP IS A STAGING DIAGRAM, NOT CONTENT: do NOT draw the\nmap itself, and do NOT draw any circles, wedges, dots, slot codes\n(E1, E2, CAM) or any other marker symbols in the frame. The output is\na clean storyboard drawing of the scene as seen by the camera only.\n\nNo text, no letters, no numbers, no captions anywhere in the image.", "reused": true, "control_asset_id": "8c882f78-6375-49aa-a388-3cce3a32f20f", "asset_id": "41c140fc-4903-4265-9a0d-2e6d22669697" }, "S9sh1": { "lane": "map_marker", "group_id": "bg_coastal_road", "segment_id": "SEG_01", "status": "ok", "image_path": "/Users/manta/Documents/Projects/TheRoad-I1/projects/ecedba66-2474-4dff-a1a4-7d13bac08e1d/images/5ea07552-480a-4fdd-a888-b6147ff239f5/conti/lane_sketch_S9sh1.png", "control_path": "/Users/manta/Documents/Projects/TheRoad-I1/projects/ecedba66-2474-4dff-a1a4-7d13bac08e1d/images/5ea07552-480a-4fdd-a888-b6147ff239f5/conti/lane_control_S9sh1.png", "base_map_path": "/Users/manta/Documents/Projects/TheRoad-I1/projects/ecedba66-2474-4dff-a1a4-7d13bac08e1d/images/background_chain/canon_bg_coastal_road_map.png", "base_map_asset_id": "a27f22eb-882a-4c81-902e-837555b3df77", "seed_path": null, "seed_asset_id": null, "geometry": { "camera": { "look_target": { "x": 0.565, "y": 0.61 }, "origin": { "x": 0.49, "y": 0.61 } }, "entity_placements": [ { "slot": "E1", "subject_en": "young woman riding a bicycle", "x": 0.565, "y": 0.61 } ], "rationale_ko": "인물을 A1 해안도로 주행 구간 중앙에 두고 남쪽으로 빠르게 달리도록 설정했다. 카메라는 같은 진행 속도로 인물의 바다 쪽 측면에서 평행 이동하며 동쪽을 바라보므로, 남쪽 진행이 화면 오른쪽 방향이 되어 엄격한 측면 자세와 페달 동작을 포착한다. B1 버스 정류장보다 아래쪽의 연속된 도로 구간을 사용해 귀가 중인 주행 순간에 집중한다." }, "leakage": { "has_marker_leakage": false, "details_ko": "스케치에 마커 기호나 평면도 요소가 없습니다." }, "prompt": "Convert the attached TOP-DOWN MARKER MAP into a single CAMERA-VIEW\nstoryboard frame (16:9), seen from the camera marked on the map.\n\nJurisdiction of sources — follow both, they do not conflict:\n- The map is the SOT for XY POSITIONS and horizontal heading ONLY.\n Each red circle labeled E1, E2, ... is where a subject stands; the\n blue dot labeled CAM is the camera position; the blue wedge is its\n horizontal viewing direction and field of view. Subjects outside\n the wedge are out of frame. Distances, directions and relative\n depths between subjects come from the map.\n- The CAMERA DIRECTION text below (when present) is the SOT for the\n VERTICAL angle, elevation, tilt and framing (eye-level, high angle,\n overhead crane, low angle, ...). If it asks for a high or overhead\n angle, draw that angle — the map being top-down does NOT mean the\n frame must be eye-level.\n\nSTYLE — LIGHT STORYBOARD FRAME: thin uniform dark outlines on plain\npaper. No shading, no hatching, no texture, no facial features beyond\na minimal head shape, no costume detail. Human figures are correctly\nproportioned full-body outlines in their canonical poses. Environment\nis reduced to a few placement lines (ground line, key landmark\nsilhouettes) — just enough to read positions and depth.\n\nMAP SLOT MEANINGS:\nE1 = young woman riding a bicycle\n\nCAMERA DIRECTION (vertical angle / elevation / framing SOT): Track parallel at torso height from 수리영’s strict side, framing slightly lower and wider so her forward-bent torso, both hands on the handlebars, bent knee, and active pedal stroke remain connected in one profile. She and the bicycle travel toward screen right across the middle of the frame, with her swept-back hair and collar extending into the trailing left side and the coastal road sliding behind her.\n\nTHE LOCATION: 해안가 도로\n\nSHOT TEXT (authoritative, Korean): 어두운 밤 해안도로를 배경으로 자전거 핸들을 꽉 쥐고 허리를 숙인 채 한쪽 무릎을 굽혀 힘차게 페달을 밟고 있는 수리영의 측면 자세, 머리카락과 옷깃이 뒤로 휘날리고 있음.\n\nCRITICAL — THE MAP IS A STAGING DIAGRAM, NOT CONTENT: do NOT draw the\nmap itself, and do NOT draw any circles, wedges, dots, slot codes\n(E1, E2, CAM) or any other marker symbols in the frame. The output is\na clean storyboard drawing of the scene as seen by the camera only.\n\nNo text, no letters, no numbers, no captions anywhere in the image.", "reused": true, "control_asset_id": "b0dc2807-f920-472e-81ba-e60616199243", "asset_id": "0cf473d2-df2c-4e0a-95be-d5edc4d0cd57" }, "S10sh3": { "lane": "structure_plate", "group_id": "bg_rooftop_residence", "segment_id": "SEG_02", "sketch_attempts": 1, "status": "ok", "image_path": "/Users/manta/Documents/Projects/TheRoad-I1/projects/ecedba66-2474-4dff-a1a4-7d13bac08e1d/images/5ea07552-480a-4fdd-a888-b6147ff239f5/conti/lane_sketch_S10sh3.png", "control_path": "/Users/manta/Documents/Projects/TheRoad-I1/projects/ecedba66-2474-4dff-a1a4-7d13bac08e1d/images/5ea07552-480a-4fdd-a888-b6147ff239f5/conti/lane_control_S10sh3.png", "base_map_path": "/Users/manta/Documents/Projects/TheRoad-I1/projects/ecedba66-2474-4dff-a1a4-7d13bac08e1d/images/background_chain/siteplan_bg_rooftop_residence.png", "base_map_asset_id": "08a2ad21-a63a-48bd-a8d8-ee6e1cb15320", "seed_path": "/Users/manta/Documents/Projects/TheRoad-I1/projects/ecedba66-2474-4dff-a1a4-7d13bac08e1d/images/background_chain/seed_bg_rooftop_residence_sel.png", "seed_asset_id": "2910d596-ddb4-4946-b782-4d6cc0e2a768", "geometry": { "camera": { "look_target": { "x": 0.505, "y": 0.315 }, "origin": { "x": 0.455, "y": 0.315 } }, "entity_placements": [ { "slot": "E1", "subject_en": "slightly ajar iron entrance door", "x": 0.505, "y": 0.315 } ], "rationale_ko": "옥상 데크(D1)에서 옥탑방 철문(D3) 바로 바깥에 카메라를 낮게 두고 문 면에 직각으로 바라보게 했다. 자물쇠(D7)와 미세한 세로 문틈이 중앙부에 놓이는 정면 정지 구도이며, 인물은 보이지 않는 순간이다." }, "geometry_attempts": 1, "leakage": { "has_marker_leakage": false, "details_ko": "도면 기호나 문자 유출 없이 건물을 그렸습니다." }, "prompt": "Convert the attached TOP-DOWN MARKER MAP into a single CAMERA-VIEW\nstoryboard frame (16:9), seen from the camera marked on the map.\n\nJurisdiction of sources — follow both, they do not conflict:\n- The map is the SOT for XY POSITIONS and horizontal heading ONLY.\n Each red circle labeled E1, E2, ... is where a subject stands; the\n blue dot labeled CAM is the camera position; the blue wedge is its\n horizontal viewing direction and field of view. Subjects outside\n the wedge are out of frame. Distances, directions and relative\n depths between subjects come from the map.\n- The CAMERA DIRECTION text below (when present) is the SOT for the\n VERTICAL angle, elevation, tilt and framing (eye-level, high angle,\n overhead crane, low angle, ...). If it asks for a high or overhead\n angle, draw that angle — the map being top-down does NOT mean the\n frame must be eye-level.\n\nSTYLE — LIGHT STORYBOARD FRAME: thin uniform dark outlines on plain\npaper. No shading, no hatching, no texture, no facial features beyond\na minimal head shape, no costume detail. Human figures are correctly\nproportioned full-body outlines in their canonical poses. Environment\nis reduced to a few placement lines (ground line, key landmark\nsilhouettes) — just enough to read positions and depth.\n\nMAP SLOT MEANINGS:\nE1 = slightly ajar iron entrance door\n\nThe SECOND attached image is the confirmed photograph of the fixed\nstructure on this site. It is the STRUCTURE LOOK authority: wherever\nthe structure appears in the sketch, its shape, proportions, openings\nand silhouette must match this photograph (rendered only as the same\nsimple line-frame style as everything else). The FIRST attached\nstaging map remains the sole authority for positions, camera placement\nand heading.\n\nCAMERA DIRECTION (vertical angle / elevation / framing SOT): Hold directly outside the entrance at low height, square to the door, with the lock hardware positioned just left of center and the hairline opening running vertically through the central area. The lock itself remains a small part of the frame while the door edges, gap, and surrounding entrance provide scale and preserve the static instant before the lateral move toward the window.\n\nTHE LOCATION: 다세대 빌라 옥상 및 옥탑방 외부\n\nSHOT TEXT (authoritative, Korean): 자물쇠가 있는 옥탑방 현관 철문이 안쪽으로 아주 미세한 틈을 벌린 채 멈춘 구도.\n\nCRITICAL — THE MAP IS A STAGING DIAGRAM, NOT CONTENT: do NOT draw the\nmap itself, and do NOT draw any circles, wedges, dots, slot codes\n(E1, E2, CAM) or any other marker symbols in the frame. The output is\na clean storyboard drawing of the scene as seen by the camera only.\n\nNo text, no letters, no numbers, no captions anywhere in the image.", "control_asset_id": "2a76f759-c82f-4101-9215-6d2a5bad2af2", "asset_id": "87f1754a-2789-455e-b726-ef68bf7f544b" }, "S10sh4": { "lane": "structure_plate", "group_id": "bg_rooftop_residence", "segment_id": "SEG_03", "sketch_attempts": 1, "status": "ok", "image_path": "/Users/manta/Documents/Projects/TheRoad-I1/projects/ecedba66-2474-4dff-a1a4-7d13bac08e1d/images/5ea07552-480a-4fdd-a888-b6147ff239f5/conti/lane_sketch_S10sh4.png", "control_path": "/Users/manta/Documents/Projects/TheRoad-I1/projects/ecedba66-2474-4dff-a1a4-7d13bac08e1d/images/5ea07552-480a-4fdd-a888-b6147ff239f5/conti/lane_control_S10sh4.png", "base_map_path": "/Users/manta/Documents/Projects/TheRoad-I1/projects/ecedba66-2474-4dff-a1a4-7d13bac08e1d/images/background_chain/siteplan_bg_rooftop_residence.png", "base_map_asset_id": "08a2ad21-a63a-48bd-a8d8-ee6e1cb15320", "seed_path": "/Users/manta/Documents/Projects/TheRoad-I1/projects/ecedba66-2474-4dff-a1a4-7d13bac08e1d/images/background_chain/seed_bg_rooftop_residence_sel.png", "seed_asset_id": "2910d596-ddb4-4946-b782-4d6cc0e2a768", "geometry": { "camera": { "look_target": { "x": 0.79, "y": 0.4 }, "origin": { "x": 0.865, "y": 0.4 } }, "entity_placements": [ { "slot": "E1", "subject_en": "intruding man lunging toward a shoulder", "x": 0.75, "y": 0.414 }, { "slot": "E2", "subject_en": "recoiling woman", "x": 0.748, "y": 0.389 } ], "rationale_ko": "D4 옥탑방 외부 창문 바로 안쪽에 두 인물을 겹쳐 배치했다. 카메라는 창과 D6 측 경계 사이의 옥상 데크에서 창면에 수직으로 서서 프레임 전체와 유리 속 실루엣을 정면으로 본다. 반사상의 좌우 반전을 고려해 실제로는 여성을 카메라 우측, 남성을 좌측에 두어 반사에서는 여성이 중간 왼쪽, 남성이 중간 오른쪽에서 어깨로 파고드는 형상이 되게 했다." }, "geometry_attempts": 1, "leakage": { "has_marker_leakage": false, "details_ko": "스케치에 마커나 기호가 보이지 않음." }, "prompt": "Convert the attached TOP-DOWN MARKER MAP into a single CAMERA-VIEW\nstoryboard frame (16:9), seen from the camera marked on the map.\n\nJurisdiction of sources — follow both, they do not conflict:\n- The map is the SOT for XY POSITIONS and horizontal heading ONLY.\n Each red circle labeled E1, E2, ... is where a subject stands; the\n blue dot labeled CAM is the camera position; the blue wedge is its\n horizontal viewing direction and field of view. Subjects outside\n the wedge are out of frame. Distances, directions and relative\n depths between subjects come from the map.\n- The CAMERA DIRECTION text below (when present) is the SOT for the\n VERTICAL angle, elevation, tilt and framing (eye-level, high angle,\n overhead crane, low angle, ...). If it asks for a high or overhead\n angle, draw that angle — the map being top-down does NOT mean the\n frame must be eye-level.\n\nSTYLE — LIGHT STORYBOARD FRAME: thin uniform dark outlines on plain\npaper. No shading, no hatching, no texture, no facial features beyond\na minimal head shape, no costume detail. Human figures are correctly\nproportioned full-body outlines in their canonical poses. Environment\nis reduced to a few placement lines (ground line, key landmark\nsilhouettes) — just enough to read positions and depth.\n\nMAP SLOT MEANINGS:\nE1 = intruding man lunging toward a shoulder\nE2 = recoiling woman\n\nThe SECOND attached image is the confirmed photograph of the fixed\nstructure on this site. It is the STRUCTURE LOOK authority: wherever\nthe structure appears in the sketch, its shape, proportions, openings\nand silhouette must match this photograph (rendered only as the same\nsimple line-frame style as everything else). The FIRST attached\nstaging map remains the sole authority for positions, camera placement\nand heading.\n\nCAMERA DIRECTION (vertical angle / elevation / framing SOT): After the lateral approach has ended, hold chest-high and perpendicular to the window so its reflected image reads without an oblique skew; the glass frame remains visible around the action. In the left-right-reversed reflection, 옥탑방 창문 너머 여성 recoils through the middle-left while 옥탑방 침입 남성 folds in from the middle-right toward her shoulder, their overlapping silhouettes frozen at the apparent bite.\n\nTHE LOCATION: 다세대 빌라 옥상 및 옥탑방 외부\n\nSHOT TEXT (authoritative, Korean): 창문 유리에 짙게 드리워진 그림자 속, 남성의 실루엣이 여성의 어깨를 거칠게 물어뜯는 듯한 역동적인 자세로 멈춘 찰나.\n\nCRITICAL — THE MAP IS A STAGING DIAGRAM, NOT CONTENT: do NOT draw the\nmap itself, and do NOT draw any circles, wedges, dots, slot codes\n(E1, E2, CAM) or any other marker symbols in the frame. The output is\na clean storyboard drawing of the scene as seen by the camera only.\n\nNo text, no letters, no numbers, no captions anywhere in the image.", "control_asset_id": "05baeb7b-abae-41c2-9e8b-6d480494fa25", "asset_id": "80db9a73-fe73-4b85-9df2-dbba20cac39d" }, "S11sh3": { "lane": "structure_plate", "group_id": "bg_rooftop_residence", "segment_id": "SEG_02", "sketch_attempts": 1, "status": "ok", "image_path": "/Users/manta/Documents/Projects/TheRoad-I1/projects/ecedba66-2474-4dff-a1a4-7d13bac08e1d/images/5ea07552-480a-4fdd-a888-b6147ff239f5/conti/lane_sketch_S11sh3.png", "control_path": "/Users/manta/Documents/Projects/TheRoad-I1/projects/ecedba66-2474-4dff-a1a4-7d13bac08e1d/images/5ea07552-480a-4fdd-a888-b6147ff239f5/conti/lane_control_S11sh3.png", "base_map_path": "/Users/manta/Documents/Projects/TheRoad-I1/projects/ecedba66-2474-4dff-a1a4-7d13bac08e1d/images/background_chain/siteplan_bg_rooftop_residence.png", "base_map_asset_id": "08a2ad21-a63a-48bd-a8d8-ee6e1cb15320", "seed_path": "/Users/manta/Documents/Projects/TheRoad-I1/projects/ecedba66-2474-4dff-a1a4-7d13bac08e1d/images/background_chain/seed_bg_rooftop_residence_sel.png", "seed_asset_id": "2910d596-ddb4-4946-b782-4d6cc0e2a768", "geometry": { "camera": { "look_target": { "x": 0.515, "y": 0.315 }, "origin": { "x": 0.705, "y": 0.675 } }, "entity_placements": [ { "slot": "E1", "subject_en": "young woman at top of stairs", "x": 0.68, "y": 0.645 } ], "rationale_ko": "E1은 외부 계단(C1)을 막 올라온 옥상 도착 지점에 두고, 카메라는 그보다 약간 뒤쪽의 어깨 높이 위치에서 철문(D3)을 바라보도록 배치했다. 인물의 몸은 프레임 밖에 남기면서 넓은 옥상 바닥(D1) 너머 상단 중앙에 반쯤 열린 철문과 어두운 내부 틈이 시선의 종착점이 되게 한다." }, "geometry_attempts": 1, "leakage": { "has_marker_leakage": false, "details_ko": "마커나 도면 기호가 발견되지 않았습니다." }, "prompt": "Convert the attached TOP-DOWN MARKER MAP into a single CAMERA-VIEW\nstoryboard frame (16:9), seen from the camera marked on the map.\n\nJurisdiction of sources — follow both, they do not conflict:\n- The map is the SOT for XY POSITIONS and horizontal heading ONLY.\n Each red circle labeled E1, E2, ... is where a subject stands; the\n blue dot labeled CAM is the camera position; the blue wedge is its\n horizontal viewing direction and field of view. Subjects outside\n the wedge are out of frame. Distances, directions and relative\n depths between subjects come from the map.\n- The CAMERA DIRECTION text below (when present) is the SOT for the\n VERTICAL angle, elevation, tilt and framing (eye-level, high angle,\n overhead crane, low angle, ...). If it asks for a high or overhead\n angle, draw that angle — the map being top-down does NOT mean the\n frame must be eye-level.\n\nSTYLE — LIGHT STORYBOARD FRAME: thin uniform dark outlines on plain\npaper. No shading, no hatching, no texture, no facial features beyond\na minimal head shape, no costume detail. Human figures are correctly\nproportioned full-body outlines in their canonical poses. Environment\nis reduced to a few placement lines (ground line, key landmark\nsilhouettes) — just enough to read positions and depth.\n\nMAP SLOT MEANINGS:\nE1 = young woman at top of stairs\n\nThe SECOND attached image is the confirmed photograph of the fixed\nstructure on this site. It is the STRUCTURE LOOK authority: wherever\nthe structure appears in the sketch, its shape, proportions, openings\nand silhouette must match this photograph (rendered only as the same\nsimple line-frame style as everything else). The FIRST attached\nstaging map remains the sole authority for positions, camera placement\nand heading.\n\nCAMERA DIRECTION (vertical angle / elevation / framing SOT): Remain at the inherited shoulder height just behind 수리영’s top-of-stairs position, but keep her body entirely outside the frame so only her eyeline across the rooftop is implied. Compose the half-open entrance in the upper-middle beyond a broad strip of rooftop floor, with the inward-facing gap exposing the dark interior as the terminal point of the shot.\n\nTHE LOCATION: 다세대 빌라 외곽 마당·주차 구역 및 옥탑방 계단\n\nSHOT TEXT (authoritative, Korean): 수리영의 시선 너머, 옥상 바닥 위로 반쯤 활짝 열린 채 내부의 어둠을 드러내고 있는 현관 철문.\n\nCRITICAL — THE MAP IS A STAGING DIAGRAM, NOT CONTENT: do NOT draw the\nmap itself, and do NOT draw any circles, wedges, dots, slot codes\n(E1, E2, CAM) or any other marker symbols in the frame. The output is\na clean storyboard drawing of the scene as seen by the camera only.\n\nNo text, no letters, no numbers, no captions anywhere in the image.", "control_asset_id": "3be2cf56-0d8d-4bfa-b70c-c416f9712d34", "asset_id": "675c0cb8-95c9-4741-9d48-47c32c700afe" }, "S13sh3": { "lane": "structure_plate", "group_id": "bg_rooftop_residence", "segment_id": "SEG_04", "sketch_attempts": 1, "status": "ok", "image_path": "/Users/manta/Documents/Projects/TheRoad-I1/projects/ecedba66-2474-4dff-a1a4-7d13bac08e1d/images/5ea07552-480a-4fdd-a888-b6147ff239f5/conti/lane_sketch_S13sh3.png", "control_path": "/Users/manta/Documents/Projects/TheRoad-I1/projects/ecedba66-2474-4dff-a1a4-7d13bac08e1d/images/5ea07552-480a-4fdd-a888-b6147ff239f5/conti/lane_control_S13sh3.png", "base_map_path": "/Users/manta/Documents/Projects/TheRoad-I1/projects/ecedba66-2474-4dff-a1a4-7d13bac08e1d/images/background_chain/siteplan_bg_rooftop_residence.png", "base_map_asset_id": "08a2ad21-a63a-48bd-a8d8-ee6e1cb15320", "seed_path": "/Users/manta/Documents/Projects/TheRoad-I1/projects/ecedba66-2474-4dff-a1a4-7d13bac08e1d/images/background_chain/seed_bg_rooftop_residence_sel.png", "seed_asset_id": "2910d596-ddb4-4946-b782-4d6cc0e2a768", "geometry": { "camera": { "look_target": { "x": 0.57, "y": 0.72 }, "origin": { "x": 0.57, "y": 0.79 } }, "entity_placements": [ { "slot": "E1", "subject_en": "dazed young woman", "x": 0.57, "y": 0.72 } ], "rationale_ko": "수리영은 후정(B3)과 외부 계단(C1)이 맞닿는 계단 하단에 멍하니 서 있도록 배치했다. 카메라는 가까운 후정 쪽에서 정적인 측면 상반신을 바라보며, 수리영 뒤로 계단이 후퇴해 보이도록 계단 축을 배경에 두었다." }, "geometry_attempts": 1, "leakage": { "has_marker_leakage": false, "details_ko": "마커나 식별 기호가 없습니다. 장면이 3D 공간으로 잘 표현되었습니다." }, "prompt": "Convert the attached TOP-DOWN MARKER MAP into a single CAMERA-VIEW\nstoryboard frame (16:9), seen from the camera marked on the map.\n\nJurisdiction of sources — follow both, they do not conflict:\n- The map is the SOT for XY POSITIONS and horizontal heading ONLY.\n Each red circle labeled E1, E2, ... is where a subject stands; the\n blue dot labeled CAM is the camera position; the blue wedge is its\n horizontal viewing direction and field of view. Subjects outside\n the wedge are out of frame. Distances, directions and relative\n depths between subjects come from the map.\n- The CAMERA DIRECTION text below (when present) is the SOT for the\n VERTICAL angle, elevation, tilt and framing (eye-level, high angle,\n overhead crane, low angle, ...). If it asks for a high or overhead\n angle, draw that angle — the map being top-down does NOT mean the\n frame must be eye-level.\n\nSTYLE — LIGHT STORYBOARD FRAME: thin uniform dark outlines on plain\npaper. No shading, no hatching, no texture, no facial features beyond\na minimal head shape, no costume detail. Human figures are correctly\nproportioned full-body outlines in their canonical poses. Environment\nis reduced to a few placement lines (ground line, key landmark\nsilhouettes) — just enough to read positions and depth.\n\nMAP SLOT MEANINGS:\nE1 = dazed young woman\n\nThe SECOND attached image is the confirmed photograph of the fixed\nstructure on this site. It is the STRUCTURE LOOK authority: wherever\nthe structure appears in the sketch, its shape, proportions, openings\nand silhouette must match this photograph (rendered only as the same\nsimple line-frame style as everything else). The FIRST attached\nstaging map remains the sole authority for positions, camera placement\nand heading.\n\nCAMERA DIRECTION (vertical angle / elevation / framing SOT): Hold a static eye-level side view near the foot of the stairs, tightened until 수리영's upper body occupies roughly two thirds of frame. Keep her blurred upward stare near the upper third and allow her lowered arms to remain visible at the bottom, with the stairs receding behind her.\n\nTHE LOCATION: 다세대 빌라 외곽 마당·주차 구역 및 옥탑방 계단\n\nSHOT TEXT (authoritative, Korean): 초점을 잃은 흐릿한 눈동자로 위쪽 허공을 응시하며 팔을 늘어뜨리고 멍하게 서 있는 수리영의 측면.\n\nCRITICAL — THE MAP IS A STAGING DIAGRAM, NOT CONTENT: do NOT draw the\nmap itself, and do NOT draw any circles, wedges, dots, slot codes\n(E1, E2, CAM) or any other marker symbols in the frame. The output is\na clean storyboard drawing of the scene as seen by the camera only.\n\nNo text, no letters, no numbers, no captions anywhere in the image.", "control_asset_id": "fe43250b-3f4d-4c90-94c2-d1c919a825b8", "asset_id": "56e6e927-36de-4c14-979c-7e72db01a9a2" }, "S13sh4": { "lane": "structure_plate", "group_id": "bg_rooftop_residence", "segment_id": "SEG_05", "sketch_attempts": 1, "status": "ok", "image_path": "/Users/manta/Documents/Projects/TheRoad-I1/projects/ecedba66-2474-4dff-a1a4-7d13bac08e1d/images/5ea07552-480a-4fdd-a888-b6147ff239f5/conti/lane_sketch_S13sh4.png", "control_path": "/Users/manta/Documents/Projects/TheRoad-I1/projects/ecedba66-2474-4dff-a1a4-7d13bac08e1d/images/5ea07552-480a-4fdd-a888-b6147ff239f5/conti/lane_control_S13sh4.png", "base_map_path": "/Users/manta/Documents/Projects/TheRoad-I1/projects/ecedba66-2474-4dff-a1a4-7d13bac08e1d/images/background_chain/siteplan_bg_rooftop_residence.png", "base_map_asset_id": "08a2ad21-a63a-48bd-a8d8-ee6e1cb15320", "seed_path": "/Users/manta/Documents/Projects/TheRoad-I1/projects/ecedba66-2474-4dff-a1a4-7d13bac08e1d/images/background_chain/seed_bg_rooftop_residence_sel.png", "seed_asset_id": "2910d596-ddb4-4946-b782-4d6cc0e2a768", "geometry": { "camera": { "look_target": { "x": 0.49, "y": 0.75 }, "origin": { "x": 0.54, "y": 0.75 } }, "entity_placements": [ { "slot": "E1", "subject_en": "shocked middle-aged woman", "x": 0.49, "y": 0.75 } ], "rationale_ko": "혜수는 하부 외부 대응 구역인 후면 안뜰(B3), 외부 계단(C1) 하단 가까이에 배치했다. 카메라는 계단 쪽에서 혜수의 정면을 향해 눈높이로 밀착해, 놀라 굳은 얼굴만 클로즈업하고 주변 인물은 프레임에서 제외한다." }, "geometry_attempts": 1, "leakage": { "has_marker_leakage": false, "details_ko": "인물의 얼굴 샷만 있으며 마커나 기호 누출은 없습니다." }, "prompt": "Convert the attached TOP-DOWN MARKER MAP into a single CAMERA-VIEW\nstoryboard frame (16:9), seen from the camera marked on the map.\n\nJurisdiction of sources — follow both, they do not conflict:\n- The map is the SOT for XY POSITIONS and horizontal heading ONLY.\n Each red circle labeled E1, E2, ... is where a subject stands; the\n blue dot labeled CAM is the camera position; the blue wedge is its\n horizontal viewing direction and field of view. Subjects outside\n the wedge are out of frame. Distances, directions and relative\n depths between subjects come from the map.\n- The CAMERA DIRECTION text below (when present) is the SOT for the\n VERTICAL angle, elevation, tilt and framing (eye-level, high angle,\n overhead crane, low angle, ...). If it asks for a high or overhead\n angle, draw that angle — the map being top-down does NOT mean the\n frame must be eye-level.\n\nSTYLE — LIGHT STORYBOARD FRAME: thin uniform dark outlines on plain\npaper. No shading, no hatching, no texture, no facial features beyond\na minimal head shape, no costume detail. Human figures are correctly\nproportioned full-body outlines in their canonical poses. Environment\nis reduced to a few placement lines (ground line, key landmark\nsilhouettes) — just enough to read positions and depth.\n\nMAP SLOT MEANINGS:\nE1 = shocked middle-aged woman\n\nThe SECOND attached image is the confirmed photograph of the fixed\nstructure on this site. It is the STRUCTURE LOOK authority: wherever\nthe structure appears in the sketch, its shape, proportions, openings\nand silhouette must match this photograph (rendered only as the same\nsimple line-frame style as everything else). The FIRST attached\nstaging map remains the sole authority for positions, camera placement\nand heading.\n\nCAMERA DIRECTION (vertical angle / elevation / framing SOT): Finish the frontal dolly-in at 혜수's eye level, framing only her widened eyes, rigid cheeks, and half-open mouth while her face occupies most of the image. The direct axis turns her involuntary shock into a brief confrontation, with all surrounding figures excluded.\n\nTHE LOCATION: 다세대 빌라 외곽 마당·주차 구역 및 옥탑방 계단\n\nSHOT TEXT (authoritative, Korean): 두 눈을 동그랗게 뜨고 입을 반쯤 벌린 채 충격으로 굳어 있는 혜수의 얼굴 클로즈업.\n\nCRITICAL — THE MAP IS A STAGING DIAGRAM, NOT CONTENT: do NOT draw the\nmap itself, and do NOT draw any circles, wedges, dots, slot codes\n(E1, E2, CAM) or any other marker symbols in the frame. The output is\na clean storyboard drawing of the scene as seen by the camera only.\n\nNo text, no letters, no numbers, no captions anywhere in the image.", "control_asset_id": "616e9099-2f4c-43a1-bc7f-b8b6dcd421c9", "asset_id": "5864857f-b543-4149-abbb-c8287e41e3b1" }, "S15sh2": { "lane": "map_marker", "group_id": "bg_coastal_road", "segment_id": "SEG_02", "status": "ok", "image_path": "/Users/manta/Documents/Projects/TheRoad-I1/projects/ecedba66-2474-4dff-a1a4-7d13bac08e1d/images/5ea07552-480a-4fdd-a888-b6147ff239f5/conti/lane_sketch_S15sh2.png", "control_path": "/Users/manta/Documents/Projects/TheRoad-I1/projects/ecedba66-2474-4dff-a1a4-7d13bac08e1d/images/5ea07552-480a-4fdd-a888-b6147ff239f5/conti/lane_control_S15sh2.png", "base_map_path": "/Users/manta/Documents/Projects/TheRoad-I1/projects/ecedba66-2474-4dff-a1a4-7d13bac08e1d/images/background_chain/canon_bg_coastal_road_map.png", "base_map_asset_id": "a27f22eb-882a-4c81-902e-837555b3df77", "seed_path": null, "seed_asset_id": null, "geometry": { "camera": { "look_target": { "x": 0.742, "y": 0.523 }, "origin": { "x": 0.742, "y": 0.548 } }, "entity_placements": [ { "slot": "E1", "subject_en": "woman extending a canned drink", "x": 0.727, "y": 0.523 }, { "slot": "E2", "subject_en": "seated young woman with folded arms", "x": 0.756, "y": 0.523 } ], "rationale_ko": "B1 해안 도로변 버스정류장의 대기 시설 앞에 두 인물을 밀착 배치했다. 캔을 내미는 여성은 도로 쪽인 왼편, 앉아 팔을 접은 젊은 여성은 정류장 시설 쪽인 오른편에 두었다. 카메라는 두 사람의 남쪽 가까이에서 가슴 높이로 중앙을 비스듬히 바라봐 왼쪽 어깨와 뻗은 팔, 중앙의 캔, 오른쪽 상체가 잡히는 근접 구도가 되도록 했다." }, "leakage": { "has_marker_leakage": false, "details_ko": "마커나 기호 유출 없음." }, "prompt": "Convert the attached TOP-DOWN MARKER MAP into a single CAMERA-VIEW\nstoryboard frame (16:9), seen from the camera marked on the map.\n\nJurisdiction of sources — follow both, they do not conflict:\n- The map is the SOT for XY POSITIONS and horizontal heading ONLY.\n Each red circle labeled E1, E2, ... is where a subject stands; the\n blue dot labeled CAM is the camera position; the blue wedge is its\n horizontal viewing direction and field of view. Subjects outside\n the wedge are out of frame. Distances, directions and relative\n depths between subjects come from the map.\n- The CAMERA DIRECTION text below (when present) is the SOT for the\n VERTICAL angle, elevation, tilt and framing (eye-level, high angle,\n overhead crane, low angle, ...). If it asks for a high or overhead\n angle, draw that angle — the map being top-down does NOT mean the\n frame must be eye-level.\n\nSTYLE — LIGHT STORYBOARD FRAME: thin uniform dark outlines on plain\npaper. No shading, no hatching, no texture, no facial features beyond\na minimal head shape, no costume detail. Human figures are correctly\nproportioned full-body outlines in their canonical poses. Environment\nis reduced to a few placement lines (ground line, key landmark\nsilhouettes) — just enough to read positions and depth.\n\nMAP SLOT MEANINGS:\nE1 = woman extending a canned drink\nE2 = seated young woman with folded arms\n\nCAMERA DIRECTION (vertical angle / elevation / framing SOT): At chest height between the two women, hold an oblique close composition with 혜수's shoulder and extended forearm entering from the left and 수리영's folded upper torso on the right. The cold can remains naturally scaled at the center, occupying only a small part of frame as it stops in the air before 수리영's chest.\n\nTHE LOCATION: 해안가 버스정류장\n\nSHOT TEXT (authoritative, Korean): 혜수의 손이 차가운 캔 음료수를 수리영의 가슴 앞쪽 허공으로 불쑥 내민 정지 순간.\n\nCRITICAL — THE MAP IS A STAGING DIAGRAM, NOT CONTENT: do NOT draw the\nmap itself, and do NOT draw any circles, wedges, dots, slot codes\n(E1, E2, CAM) or any other marker symbols in the frame. The output is\na clean storyboard drawing of the scene as seen by the camera only.\n\nNo text, no letters, no numbers, no captions anywhere in the image.", "reused": true, "control_asset_id": "bbe497b9-f558-4ed7-8b31-b2686fb8878d", "asset_id": "3364d8d2-a48b-4d3f-bcfd-7b465b4a789f" }, "S15sh5": { "lane": "map_marker", "group_id": "bg_coastal_road", "segment_id": "SEG_02", "status": "ok", "image_path": "/Users/manta/Documents/Projects/TheRoad-I1/projects/ecedba66-2474-4dff-a1a4-7d13bac08e1d/images/5ea07552-480a-4fdd-a888-b6147ff239f5/conti/lane_sketch_S15sh5.png", "control_path": "/Users/manta/Documents/Projects/TheRoad-I1/projects/ecedba66-2474-4dff-a1a4-7d13bac08e1d/images/5ea07552-480a-4fdd-a888-b6147ff239f5/conti/lane_control_S15sh5.png", "base_map_path": "/Users/manta/Documents/Projects/TheRoad-I1/projects/ecedba66-2474-4dff-a1a4-7d13bac08e1d/images/background_chain/canon_bg_coastal_road_map.png", "base_map_asset_id": "a27f22eb-882a-4c81-902e-837555b3df77", "seed_path": null, "seed_asset_id": null, "geometry": { "camera": { "look_target": { "x": 0.64, "y": 0.49 }, "origin": { "x": 0.49, "y": 0.58 } }, "entity_placements": [ { "slot": "E1", "subject_en": "running young woman with one leg lifted", "x": 0.59, "y": 0.47 }, { "slot": "E2", "subject_en": "standing woman calling with hands cupped around mouth", "x": 0.7, "y": 0.55 } ], "rationale_ko": "수리영은 B1 버스정류장에서 벗어나 A1 해안도로를 따라 북서쪽으로 달려가는 위치에 두고, 혜수는 정류장에 남겨 두 사람 사이의 거리를 확보했다. 카메라는 주행 경로의 남서쪽에 두어 수리영의 들린 다리와 뒷모습이 좌측 전경을 차지하고, 입가에 양손을 모은 혜수의 상체가 우측 원경에 작게 보이도록 정렬했다. 수리영의 진행 방향 앞쪽에는 해안도로의 열린 공간을 남겼다." }, "leakage": { "has_marker_leakage": false, "details_ko": "마커나 기호 유출 없음." }, "prompt": "Convert the attached TOP-DOWN MARKER MAP into a single CAMERA-VIEW\nstoryboard frame (16:9), seen from the camera marked on the map.\n\nJurisdiction of sources — follow both, they do not conflict:\n- The map is the SOT for XY POSITIONS and horizontal heading ONLY.\n Each red circle labeled E1, E2, ... is where a subject stands; the\n blue dot labeled CAM is the camera position; the blue wedge is its\n horizontal viewing direction and field of view. Subjects outside\n the wedge are out of frame. Distances, directions and relative\n depths between subjects come from the map.\n- The CAMERA DIRECTION text below (when present) is the SOT for the\n VERTICAL angle, elevation, tilt and framing (eye-level, high angle,\n overhead crane, low angle, ...). If it asks for a high or overhead\n angle, draw that angle — the map being top-down does NOT mean the\n frame must be eye-level.\n\nSTYLE — LIGHT STORYBOARD FRAME: thin uniform dark outlines on plain\npaper. No shading, no hatching, no texture, no facial features beyond\na minimal head shape, no costume detail. Human figures are correctly\nproportioned full-body outlines in their canonical poses. Environment\nis reduced to a few placement lines (ground line, key landmark\nsilhouettes) — just enough to read positions and depth.\n\nMAP SLOT MEANINGS:\nE1 = running young woman with one leg lifted\nE2 = standing woman calling with hands cupped around mouth\n\nCAMERA DIRECTION (vertical angle / elevation / framing SOT): From upper-body height beside the running path, hold a static wide alignment across 수리영's rear: her lifted leg and running back occupy the left foreground while 혜수's calling upper body remains small in the right background. The detached, watchful framing preserves clear distance between them and leaves open space ahead of 수리영's movement.\n\nTHE LOCATION: 해안가 버스정류장\n\nSHOT TEXT (authoritative, Korean): 거리를 둔 채 한쪽 다리를 들고 달리는 자세인 수리영의 뒷모습 너머, 입가에 양손을 나팔 모양으로 모으고 입을 벌린 채 서 있는 혜수의 상체가 원경으로 보이는 구도.\n\nCRITICAL — THE MAP IS A STAGING DIAGRAM, NOT CONTENT: do NOT draw the\nmap itself, and do NOT draw any circles, wedges, dots, slot codes\n(E1, E2, CAM) or any other marker symbols in the frame. The output is\na clean storyboard drawing of the scene as seen by the camera only.\n\nNo text, no letters, no numbers, no captions anywhere in the image.", "reused": true, "control_asset_id": "1b974707-6ca9-405d-bde4-051d4e3a98a5", "asset_id": "5459008c-b62f-408f-8a56-0f0a7731af7c" }, "S17sh1": { "lane": "structure_plate", "group_id": "bg_rooftop_residence", "segment_id": "SEG_03", "sketch_attempts": 1, "status": "ok", "image_path": "/Users/manta/Documents/Projects/TheRoad-I1/projects/ecedba66-2474-4dff-a1a4-7d13bac08e1d/images/5ea07552-480a-4fdd-a888-b6147ff239f5/conti/lane_sketch_S17sh1.png", "control_path": "/Users/manta/Documents/Projects/TheRoad-I1/projects/ecedba66-2474-4dff-a1a4-7d13bac08e1d/images/5ea07552-480a-4fdd-a888-b6147ff239f5/conti/lane_control_S17sh1.png", "base_map_path": "/Users/manta/Documents/Projects/TheRoad-I1/projects/ecedba66-2474-4dff-a1a4-7d13bac08e1d/images/background_chain/siteplan_bg_rooftop_residence.png", "base_map_asset_id": "08a2ad21-a63a-48bd-a8d8-ee6e1cb15320", "seed_path": "/Users/manta/Documents/Projects/TheRoad-I1/projects/ecedba66-2474-4dff-a1a4-7d13bac08e1d/images/background_chain/seed_bg_rooftop_residence_sel.png", "seed_asset_id": "2910d596-ddb4-4946-b782-4d6cc0e2a768", "geometry": { "camera": { "look_target": { "x": 0.78, "y": 0.28 }, "origin": { "x": 0.87, "y": 0.61 } }, "entity_placements": [ { "slot": "E1", "subject_en": "black cat standing on boundary wall", "x": 0.76, "y": 0.15 } ], "rationale_ko": "검은 고양이는 옥탑방 위쪽의 경계 담장(D6) 상단에 네 발로 서고, 오른쪽 벽의 창문(D4)을 향해 고개를 돌릴 수 있도록 배치했다. 카메라는 옥상 데크(D1) 오른쪽 아래에서 고양이 몸높이로 북서쪽을 바라보며, 고양이가 화면 왼쪽 1/3에 놓이고 불 켜진 창문이 그 시선 방향의 오른쪽에 함께 들어오는 넓은 3/4 구도를 형성한다." }, "geometry_attempts": 1, "leakage": { "has_marker_leakage": false, "details_ko": "도면 마커나 문자 기호가 발견되지 않았습니다." }, "prompt": "Convert the attached TOP-DOWN MARKER MAP into a single CAMERA-VIEW\nstoryboard frame (16:9), seen from the camera marked on the map.\n\nJurisdiction of sources — follow both, they do not conflict:\n- The map is the SOT for XY POSITIONS and horizontal heading ONLY.\n Each red circle labeled E1, E2, ... is where a subject stands; the\n blue dot labeled CAM is the camera position; the blue wedge is its\n horizontal viewing direction and field of view. Subjects outside\n the wedge are out of frame. Distances, directions and relative\n depths between subjects come from the map.\n- The CAMERA DIRECTION text below (when present) is the SOT for the\n VERTICAL angle, elevation, tilt and framing (eye-level, high angle,\n overhead crane, low angle, ...). If it asks for a high or overhead\n angle, draw that angle — the map being top-down does NOT mean the\n frame must be eye-level.\n\nSTYLE — LIGHT STORYBOARD FRAME: thin uniform dark outlines on plain\npaper. No shading, no hatching, no texture, no facial features beyond\na minimal head shape, no costume detail. Human figures are correctly\nproportioned full-body outlines in their canonical poses. Environment\nis reduced to a few placement lines (ground line, key landmark\nsilhouettes) — just enough to read positions and depth.\n\nMAP SLOT MEANINGS:\nE1 = black cat standing on boundary wall\n\nThe SECOND attached image is the confirmed photograph of the fixed\nstructure on this site. It is the STRUCTURE LOOK authority: wherever\nthe structure appears in the sketch, its shape, proportions, openings\nand silhouette must match this photograph (rendered only as the same\nsimple line-frame style as everything else). The FIRST attached\nstaging map remains the sole authority for positions, camera placement\nand heading.\n\nCAMERA DIRECTION (vertical angle / elevation / framing SOT): At the black cat's body height and several meters from the wall, hold a static wide three-quarter view with the cat silhouetted on the left third and the lit rooftop-room window set along its sightline in the upper right. Keep the wall's top line connecting both anchors, pausing at the instant the cat turns its head toward the window before the lateral track begins.\n\nTHE LOCATION: 다세대 빌라 옥상 및 옥탑방 외부\n\nSHOT TEXT (authoritative, Korean): 여명이 밝아오는 푸르스름한 새벽빛 아래, 불 켜진 옥탑방 창문을 향해 고개를 돌린 검은 고양이가 시멘트 담벼락 위에 네 발로 서 있는 실루엣.\n\nCRITICAL — THE MAP IS A STAGING DIAGRAM, NOT CONTENT: do NOT draw the\nmap itself, and do NOT draw any circles, wedges, dots, slot codes\n(E1, E2, CAM) or any other marker symbols in the frame. The output is\na clean storyboard drawing of the scene as seen by the camera only.\n\nNo text, no letters, no numbers, no captions anywhere in the image.", "control_asset_id": "892c5415-cd07-4e8e-9ee3-0aa1c52034d0", "asset_id": "2d593d6f-c492-4e06-b0b4-c612eb12d958" }, "S22sh1": { "lane": "structure_plate", "group_id": "bg_coastal_road", "segment_id": "SEG_01", "sketch_attempts": 1, "status": "ok", "image_path": "/Users/manta/Documents/Projects/TheRoad-I1/projects/ecedba66-2474-4dff-a1a4-7d13bac08e1d/images/5ea07552-480a-4fdd-a888-b6147ff239f5/conti/lane_sketch_S22sh1.png", "control_path": "/Users/manta/Documents/Projects/TheRoad-I1/projects/ecedba66-2474-4dff-a1a4-7d13bac08e1d/images/5ea07552-480a-4fdd-a888-b6147ff239f5/conti/lane_control_S22sh1.png", "base_map_path": "/Users/manta/Documents/Projects/TheRoad-I1/projects/ecedba66-2474-4dff-a1a4-7d13bac08e1d/images/background_chain/siteplan_bg_coastal_road.png", "base_map_asset_id": "94882417-dd76-46ab-b2a3-0e222da52310", "seed_path": "/Users/manta/Documents/Projects/TheRoad-I1/projects/ecedba66-2474-4dff-a1a4-7d13bac08e1d/images/background_chain/seed_bg_coastal_road_sel.png", "seed_asset_id": "324395f4-57ad-4523-96af-a9c4c1d824a4", "geometry": { "camera": { "look_target": { "x": 0.55, "y": 0.52 }, "origin": { "x": 0.45, "y": 0.6 } }, "entity_placements": [ { "slot": "E1", "subject_en": "young woman riding a bicycle", "x": 0.52, "y": 0.49 } ], "rationale_ko": "자전거를 탄 인물을 A1 해안 도로 주행로 중앙에 두고 도로를 따라 화면 오른쪽에 해당하는 남동 방향으로 달리게 배치했다. 카메라는 인물의 육지 쪽 측면에서 자전거 진행축과 평행하게 이동하도록 두었으며, 시선을 진행 방향보다 약간 앞에 맞춰 인물이 프레임 중간 왼쪽에 놓이고 굽힌 무릎과 페달 동작, 뒤로 흩날리는 머리카락이 온전히 보이도록 했다." }, "geometry_attempts": 1, "leakage": { "has_marker_leakage": false, "details_ko": "제시된 스케치에는 카메라 마커나 문자 등의 누출이 없습니다." }, "prompt": "Convert the attached TOP-DOWN MARKER MAP into a single CAMERA-VIEW\nstoryboard frame (16:9), seen from the camera marked on the map.\n\nJurisdiction of sources — follow both, they do not conflict:\n- The map is the SOT for XY POSITIONS and horizontal heading ONLY.\n Each red circle labeled E1, E2, ... is where a subject stands; the\n blue dot labeled CAM is the camera position; the blue wedge is its\n horizontal viewing direction and field of view. Subjects outside\n the wedge are out of frame. Distances, directions and relative\n depths between subjects come from the map.\n- The CAMERA DIRECTION text below (when present) is the SOT for the\n VERTICAL angle, elevation, tilt and framing (eye-level, high angle,\n overhead crane, low angle, ...). If it asks for a high or overhead\n angle, draw that angle — the map being top-down does NOT mean the\n frame must be eye-level.\n\nSTYLE — LIGHT STORYBOARD FRAME: thin uniform dark outlines on plain\npaper. No shading, no hatching, no texture, no facial features beyond\na minimal head shape, no costume detail. Human figures are correctly\nproportioned full-body outlines in their canonical poses. Environment\nis reduced to a few placement lines (ground line, key landmark\nsilhouettes) — just enough to read positions and depth.\n\nMAP SLOT MEANINGS:\nE1 = young woman riding a bicycle\n\nThe SECOND attached image is the confirmed photograph of the fixed\nstructure on this site. It is the STRUCTURE LOOK authority: wherever\nthe structure appears in the sketch, its shape, proportions, openings\nand silhouette must match this photograph (rendered only as the same\nsimple line-frame style as everything else). The FIRST attached\nstaging map remains the sole authority for positions, camera placement\nand heading.\n\nCAMERA DIRECTION (vertical angle / elevation / framing SOT): At torso height and parallel to the bicycle, track alongside 수리영 from the specified lateral distance while framing her full riding action in profile. She occupies the middle-left area moving toward screen right, with the bent knee and downward pedal stroke held near frame center and her wind-swept hair trailing into the negative space behind.\n\nTHE LOCATION: 해안가 도로\n\nSHOT TEXT (authoritative, Korean): 밝은 낮, 야외에서 자전거 안장에 앉아 한쪽 무릎을 굽혀 페달을 밟아 내리는 주행 중간 동작으로, 머리카락이 바람에 뒤로 흩날리는 수리영의 측면 모습.\n\nCRITICAL — THE MAP IS A STAGING DIAGRAM, NOT CONTENT: do NOT draw the\nmap itself, and do NOT draw any circles, wedges, dots, slot codes\n(E1, E2, CAM) or any other marker symbols in the frame. The output is\na clean storyboard drawing of the scene as seen by the camera only.\n\nNo text, no letters, no numbers, no captions anywhere in the image.", "control_asset_id": "b4a9daa8-fcbc-4faf-972e-d043a0c9244c", "asset_id": "fb3ab841-c216-4853-b7ec-40486e7df1e5" }, "S24sh1": { "lane": "map_marker", "group_id": "bg_coastal_road", "segment_id": "SEG_01", "status": "ok", "image_path": "/Users/manta/Documents/Projects/TheRoad-I1/projects/ecedba66-2474-4dff-a1a4-7d13bac08e1d/images/5ea07552-480a-4fdd-a888-b6147ff239f5/conti/lane_sketch_S24sh1.png", "control_path": "/Users/manta/Documents/Projects/TheRoad-I1/projects/ecedba66-2474-4dff-a1a4-7d13bac08e1d/images/5ea07552-480a-4fdd-a888-b6147ff239f5/conti/lane_control_S24sh1.png", "base_map_path": "/Users/manta/Documents/Projects/TheRoad-I1/projects/ecedba66-2474-4dff-a1a4-7d13bac08e1d/images/background_chain/canon_bg_coastal_road_map.png", "base_map_asset_id": "a27f22eb-882a-4c81-902e-837555b3df77", "seed_path": null, "seed_asset_id": null, "geometry": { "camera": { "look_target": { "x": 0.57, "y": 0.58 }, "origin": { "x": 0.48, "y": 0.58 } }, "entity_placements": [ { "slot": "E1", "subject_en": "young woman riding a bicycle", "x": 0.57, "y": 0.58 } ], "rationale_ko": "자전거를 탄 인물을 버스 정류장(B1) 옆으로 이어지는 해안 도로 회랑(A1)의 주행 차로에 배치했다. 인물은 도로를 남쪽으로 달리는 순간으로 설정하고, 카메라는 해안 쪽에서 같은 속도로 병행 이동하며 동쪽을 바라보게 하여 전신과 자전거 전체가 보이는 넓은 측면 구도를 만든다." }, "leakage": { "has_marker_leakage": false, "details_ko": "마커나 기호가 보이지 않음." }, "prompt": "Convert the attached TOP-DOWN MARKER MAP into a single CAMERA-VIEW\nstoryboard frame (16:9), seen from the camera marked on the map.\n\nJurisdiction of sources — follow both, they do not conflict:\n- The map is the SOT for XY POSITIONS and horizontal heading ONLY.\n Each red circle labeled E1, E2, ... is where a subject stands; the\n blue dot labeled CAM is the camera position; the blue wedge is its\n horizontal viewing direction and field of view. Subjects outside\n the wedge are out of frame. Distances, directions and relative\n depths between subjects come from the map.\n- The CAMERA DIRECTION text below (when present) is the SOT for the\n VERTICAL angle, elevation, tilt and framing (eye-level, high angle,\n overhead crane, low angle, ...). If it asks for a high or overhead\n angle, draw that angle — the map being top-down does NOT mean the\n frame must be eye-level.\n\nSTYLE — LIGHT STORYBOARD FRAME: thin uniform dark outlines on plain\npaper. No shading, no hatching, no texture, no facial features beyond\na minimal head shape, no costume detail. Human figures are correctly\nproportioned full-body outlines in their canonical poses. Environment\nis reduced to a few placement lines (ground line, key landmark\nsilhouettes) — just enough to read positions and depth.\n\nMAP SLOT MEANINGS:\nE1 = young woman riding a bicycle\n\nCAMERA DIRECTION (vertical angle / elevation / framing SOT): Track laterally beside 수리영 at bicycle height, holding a wide profile that contains her full body and the entire bicycle as the driven pedal reaches the bottom of its stroke. Keep her silhouette centered slightly forward in the frame, leaving space ahead of the bicycle while her clothing and the rain scatter backward across the trailing side.\n\nTHE LOCATION: 해안가 도로\n\nSHOT TEXT (authoritative, Korean): 짙은 먹구름과 세찬 비가 쏟아지는 해안가 도로 위, 빗물에 젖은 수리영이 자전거 안장에 앉아 한쪽 페달을 힘껏 밟아 누른 채 핸들을 꽉 쥔 전신 실루엣, 옷자락과 빗방울이 뒤로 흩날리는 찰나.\n\nCRITICAL — THE MAP IS A STAGING DIAGRAM, NOT CONTENT: do NOT draw the\nmap itself, and do NOT draw any circles, wedges, dots, slot codes\n(E1, E2, CAM) or any other marker symbols in the frame. The output is\na clean storyboard drawing of the scene as seen by the camera only.\n\nNo text, no letters, no numbers, no captions anywhere in the image.", "reused": true, "control_asset_id": "5386caf5-7cfd-49d7-b2c4-eab9cd89a64f", "asset_id": "1aa22a46-0ed2-45e7-8462-346ce7fc49eb" }, "S26sh2": { "lane": "map_marker", "group_id": "bg_foggy_harbor", "segment_id": "SEG_01", "status": "ok", "image_path": "/Users/manta/Documents/Projects/TheRoad-I1/projects/ecedba66-2474-4dff-a1a4-7d13bac08e1d/images/5ea07552-480a-4fdd-a888-b6147ff239f5/conti/lane_sketch_S26sh2.png", "control_path": "/Users/manta/Documents/Projects/TheRoad-I1/projects/ecedba66-2474-4dff-a1a4-7d13bac08e1d/images/5ea07552-480a-4fdd-a888-b6147ff239f5/conti/lane_control_S26sh2.png", "base_map_path": "/Users/manta/Documents/Projects/TheRoad-I1/projects/ecedba66-2474-4dff-a1a4-7d13bac08e1d/images/background_chain/canon_bg_foggy_harbor_map.png", "base_map_asset_id": "bdc24701-e9db-4bd8-a371-0a33662587af", "seed_path": null, "seed_asset_id": null, "geometry": { "camera": { "look_target": { "x": 0.27, "y": 0.48 }, "origin": { "x": 0.27, "y": 0.56 } }, "entity_placements": [ { "slot": "E1", "subject_en": "seated young woman with one knee raised", "x": 0.27, "y": 0.48 } ], "rationale_ko": "인물은 R1의 앉을 수 있는 젖은 갯바위 군집에 배치하고, 오른쪽의 H1과 R2 방향 안개 낀 바다를 바라보게 설정했다. 카메라는 인물의 남쪽 인접 갯바위에서 시선축과 직각으로 측면을 포착하며, 바다 쪽 화면 공간이 열리는 정적인 미디엄 프로필 구도다." }, "leakage": { "has_marker_leakage": false, "details_ko": "마커나 기호 누출 없이 정상적으로 그려진 스케치입니다." }, "prompt": "Convert the attached TOP-DOWN MARKER MAP into a single CAMERA-VIEW\nstoryboard frame (16:9), seen from the camera marked on the map.\n\nJurisdiction of sources — follow both, they do not conflict:\n- The map is the SOT for XY POSITIONS and horizontal heading ONLY.\n Each red circle labeled E1, E2, ... is where a subject stands; the\n blue dot labeled CAM is the camera position; the blue wedge is its\n horizontal viewing direction and field of view. Subjects outside\n the wedge are out of frame. Distances, directions and relative\n depths between subjects come from the map.\n- The CAMERA DIRECTION text below (when present) is the SOT for the\n VERTICAL angle, elevation, tilt and framing (eye-level, high angle,\n overhead crane, low angle, ...). If it asks for a high or overhead\n angle, draw that angle — the map being top-down does NOT mean the\n frame must be eye-level.\n\nSTYLE — LIGHT STORYBOARD FRAME: thin uniform dark outlines on plain\npaper. No shading, no hatching, no texture, no facial features beyond\na minimal head shape, no costume detail. Human figures are correctly\nproportioned full-body outlines in their canonical poses. Environment\nis reduced to a few placement lines (ground line, key landmark\nsilhouettes) — just enough to read positions and depth.\n\nMAP SLOT MEANINGS:\nE1 = seated young woman with one knee raised\n\nCAMERA DIRECTION (vertical angle / elevation / framing SOT): Set the camera beside the rocks at 수리영's seated shoulder height, perpendicular to her sightline, and hold a static medium profile against the fog-covered sea. Place her in one lateral third with a raised knee breaking the lower silhouette, leaving the opposite side open toward the water she watches before the camera descends toward the phone.\n\nTHE LOCATION: 안개 낀 포구와 갯바위\n\nSHOT TEXT (authoritative, Korean): 안개 낀 바다를 향해 젖은 갯바위 위에 무릎을 세우고 앉은 수리영의 측면.\n\nCRITICAL — THE MAP IS A STAGING DIAGRAM, NOT CONTENT: do NOT draw the\nmap itself, and do NOT draw any circles, wedges, dots, slot codes\n(E1, E2, CAM) or any other marker symbols in the frame. The output is\na clean storyboard drawing of the scene as seen by the camera only.\n\nNo text, no letters, no numbers, no captions anywhere in the image.", "reused": true, "control_asset_id": "65622eb8-9cf7-4d2c-8b2b-7de5ff954714", "asset_id": "6a72ff5b-3d6e-4d5d-9c74-3f4b4898b8af" }, "S26sh6": { "lane": "map_marker", "group_id": "bg_foggy_harbor", "segment_id": "SEG_02", "status": "ok", "image_path": "/Users/manta/Documents/Projects/TheRoad-I1/projects/ecedba66-2474-4dff-a1a4-7d13bac08e1d/images/5ea07552-480a-4fdd-a888-b6147ff239f5/conti/lane_sketch_S26sh6.png", "control_path": "/Users/manta/Documents/Projects/TheRoad-I1/projects/ecedba66-2474-4dff-a1a4-7d13bac08e1d/images/5ea07552-480a-4fdd-a888-b6147ff239f5/conti/lane_control_S26sh6.png", "base_map_path": "/Users/manta/Documents/Projects/TheRoad-I1/projects/ecedba66-2474-4dff-a1a4-7d13bac08e1d/images/background_chain/canon_bg_foggy_harbor_map.png", "base_map_asset_id": "bdc24701-e9db-4bd8-a371-0a33662587af", "seed_path": null, "seed_asset_id": null, "geometry": { "camera": { "look_target": { "x": 0.606, "y": 0.324 }, "origin": { "x": 0.657, "y": 0.365 } }, "entity_placements": [ { "slot": "E1", "subject_en": "young woman standing with chin lifted", "x": 0.629, "y": 0.347 }, { "slot": "E2", "subject_en": "young man standing above near the moored boat", "x": 0.604, "y": 0.322 } ], "rationale_ko": "H3 선착장 계류점 바로 앞의 접근 가능한 랜딩 바닥에 여성을 두고, 남성은 계류된 배에 가까운 물가 쪽 가장자리에 배치해 두 사람의 높이 차와 마주 보는 시선을 형성했다. 카메라는 여성의 옆으로 몇 걸음 떨어진 랜딩 바닥에 두고 H3 방향으로 촬영해 여성이 전경, 남성이 상부 배경에 놓이도록 했다." }, "leakage": { "has_marker_leakage": false, "details_ko": "도면 마커나 기호 노출이 없습니다." }, "prompt": "Convert the attached TOP-DOWN MARKER MAP into a single CAMERA-VIEW\nstoryboard frame (16:9), seen from the camera marked on the map.\n\nJurisdiction of sources — follow both, they do not conflict:\n- The map is the SOT for XY POSITIONS and horizontal heading ONLY.\n Each red circle labeled E1, E2, ... is where a subject stands; the\n blue dot labeled CAM is the camera position; the blue wedge is its\n horizontal viewing direction and field of view. Subjects outside\n the wedge are out of frame. Distances, directions and relative\n depths between subjects come from the map.\n- The CAMERA DIRECTION text below (when present) is the SOT for the\n VERTICAL angle, elevation, tilt and framing (eye-level, high angle,\n overhead crane, low angle, ...). If it asks for a high or overhead\n angle, draw that angle — the map being top-down does NOT mean the\n frame must be eye-level.\n\nSTYLE — LIGHT STORYBOARD FRAME: thin uniform dark outlines on plain\npaper. No shading, no hatching, no texture, no facial features beyond\na minimal head shape, no costume detail. Human figures are correctly\nproportioned full-body outlines in their canonical poses. Environment\nis reduced to a few placement lines (ground line, key landmark\nsilhouettes) — just enough to read positions and depth.\n\nMAP SLOT MEANINGS:\nE1 = young woman standing with chin lifted\nE2 = young man standing above near the moored boat\n\nCAMERA DIRECTION (vertical angle / elevation / framing SOT): Place the camera at low waist height on the ground immediately before the boat, several feet to 수리영's side, and angle upward to hold both levels of the exchange. 수리영 occupies the lower foreground with her chin lifted, while 인우 remains above her near the boat in the upper background, their opposing eyelines joining across the frame.\n\nTHE LOCATION: 안개 낀 포구와 갯바위\n\nSHOT TEXT (authoritative, Korean): 포구에 닿은 배 바로 앞 바닥에 선 수리영이 위쪽의 인우를 향해 고개를 든 정지 상태.\n\nCRITICAL — THE MAP IS A STAGING DIAGRAM, NOT CONTENT: do NOT draw the\nmap itself, and do NOT draw any circles, wedges, dots, slot codes\n(E1, E2, CAM) or any other marker symbols in the frame. The output is\na clean storyboard drawing of the scene as seen by the camera only.\n\nNo text, no letters, no numbers, no captions anywhere in the image.", "reused": true, "control_asset_id": "ff9832c4-7e12-4039-8517-17e20990a4ad", "asset_id": "876f9bef-8fcc-44d3-95a7-9b364afd99c7" }, "S28sh3": { "lane": "map_marker", "group_id": "bg_small_fishing_boat", "segment_id": "SEG_01", "status": "ok", "image_path": "/Users/manta/Documents/Projects/TheRoad-I1/projects/ecedba66-2474-4dff-a1a4-7d13bac08e1d/images/5ea07552-480a-4fdd-a888-b6147ff239f5/conti/lane_sketch_S28sh3.png", "control_path": "/Users/manta/Documents/Projects/TheRoad-I1/projects/ecedba66-2474-4dff-a1a4-7d13bac08e1d/images/5ea07552-480a-4fdd-a888-b6147ff239f5/conti/lane_control_S28sh3.png", "base_map_path": "/Users/manta/Documents/Projects/TheRoad-I1/projects/ecedba66-2474-4dff-a1a4-7d13bac08e1d/images/background_chain/canon_bg_small_fishing_boat_map.png", "base_map_asset_id": "344cb211-e88c-420d-b314-231478857b60", "seed_path": null, "seed_asset_id": null, "geometry": { "camera": { "look_target": { "x": 0.476, "y": 0.402 }, "origin": { "x": 0.438, "y": 0.414 } }, "entity_placements": [ { "slot": "E1", "subject_en": "young woman raising an old photograph", "x": 0.478, "y": 0.404 }, { "slot": "E2", "subject_en": "seated man watching the photograph", "x": 0.545, "y": 0.374 } ], "rationale_ko": "두 인물을 D1 갑판의 W2 조타실 입구 앞 열린 동선에 배치했다. 카메라는 여성의 치켜든 손 바로 옆 클로즈업 거리에 두고 사진을 정면으로 바라보며, 뒤쪽 갑판과 남성은 흐릿한 배경으로 남도록 했다." }, "leakage": { "has_marker_leakage": false, "details_ko": "마커나 기호가 발견되지 않았습니다." }, "prompt": "Convert the attached TOP-DOWN MARKER MAP into a single CAMERA-VIEW\nstoryboard frame (16:9), seen from the camera marked on the map.\n\nJurisdiction of sources — follow both, they do not conflict:\n- The map is the SOT for XY POSITIONS and horizontal heading ONLY.\n Each red circle labeled E1, E2, ... is where a subject stands; the\n blue dot labeled CAM is the camera position; the blue wedge is its\n horizontal viewing direction and field of view. Subjects outside\n the wedge are out of frame. Distances, directions and relative\n depths between subjects come from the map.\n- The CAMERA DIRECTION text below (when present) is the SOT for the\n VERTICAL angle, elevation, tilt and framing (eye-level, high angle,\n overhead crane, low angle, ...). If it asks for a high or overhead\n angle, draw that angle — the map being top-down does NOT mean the\n frame must be eye-level.\n\nSTYLE — LIGHT STORYBOARD FRAME: thin uniform dark outlines on plain\npaper. No shading, no hatching, no texture, no facial features beyond\na minimal head shape, no costume detail. Human figures are correctly\nproportioned full-body outlines in their canonical poses. Environment\nis reduced to a few placement lines (ground line, key landmark\nsilhouettes) — just enough to read positions and depth.\n\nMAP SLOT MEANINGS:\nE1 = young woman raising an old photograph\nE2 = seated man watching the photograph\n\nCAMERA DIRECTION (vertical angle / elevation / framing SOT): Hold at 수리영's hand height within close-up distance, with the raised photograph occupying less than two-fifths of the central frame and her pinching fingers supplying a human scale reference. Keep the photograph's image face square enough for the small writing '금월' to remain readable, while partial deck context and 인우's indistinct position remain beyond it before the lateral move begins.\n\nTHE LOCATION: 소형 어선 갑판\n\nSHOT TEXT (authoritative, Korean): 수리영의 한 손이 '금월'이라는 작은 글씨가 적힌 낡은 사진을 공중에 치켜든 클로즈업.\n\nCRITICAL — THE MAP IS A STAGING DIAGRAM, NOT CONTENT: do NOT draw the\nmap itself, and do NOT draw any circles, wedges, dots, slot codes\n(E1, E2, CAM) or any other marker symbols in the frame. The output is\na clean storyboard drawing of the scene as seen by the camera only.\n\nNo text, no letters, no numbers, no captions anywhere in the image.", "reused": true, "control_asset_id": "2e9c0ccc-c4ee-4e3b-b10f-8b3733bc1fb7", "asset_id": "ab5e5a06-fdca-4e39-ac7f-b6a6006ee359" }, "S28sh4": { "lane": "map_marker", "group_id": "bg_small_fishing_boat", "segment_id": "SEG_01", "status": "ok", "image_path": "/Users/manta/Documents/Projects/TheRoad-I1/projects/ecedba66-2474-4dff-a1a4-7d13bac08e1d/images/5ea07552-480a-4fdd-a888-b6147ff239f5/conti/lane_sketch_S28sh4.png", "control_path": "/Users/manta/Documents/Projects/TheRoad-I1/projects/ecedba66-2474-4dff-a1a4-7d13bac08e1d/images/5ea07552-480a-4fdd-a888-b6147ff239f5/conti/lane_control_S28sh4.png", "base_map_path": "/Users/manta/Documents/Projects/TheRoad-I1/projects/ecedba66-2474-4dff-a1a4-7d13bac08e1d/images/background_chain/canon_bg_small_fishing_boat_map.png", "base_map_asset_id": "344cb211-e88c-420d-b314-231478857b60", "seed_path": null, "seed_asset_id": null, "geometry": { "camera": { "look_target": { "x": 0.5, "y": 0.405 }, "origin": { "x": 0.455, "y": 0.35 } }, "entity_placements": [ { "slot": "E1", "subject_en": "frozen seated man", "x": 0.5, "y": 0.405 } ], "rationale_ko": "인우를 D1 갑판에서 W2 조타실 입구 바로 앞쪽에 배치해, 뒤로 일어나 곧장 조타실로 향할 수 있는 동선을 확보했다. 카메라는 사진이 있던 전방 위치를 옆으로 지난 뒤 인우의 상흉부 높이에서 가까이 접근한 근정면 3/4 방향이며, 인우 뒤쪽으로 W2를 향한 짧은 여백이 남도록 설정했다." }, "leakage": { "has_marker_leakage": false, "details_ko": "도면 마커나 평면도 기호가 없음." }, "prompt": "Convert the attached TOP-DOWN MARKER MAP into a single CAMERA-VIEW\nstoryboard frame (16:9), seen from the camera marked on the map.\n\nJurisdiction of sources — follow both, they do not conflict:\n- The map is the SOT for XY POSITIONS and horizontal heading ONLY.\n Each red circle labeled E1, E2, ... is where a subject stands; the\n blue dot labeled CAM is the camera position; the blue wedge is its\n horizontal viewing direction and field of view. Subjects outside\n the wedge are out of frame. Distances, directions and relative\n depths between subjects come from the map.\n- The CAMERA DIRECTION text below (when present) is the SOT for the\n VERTICAL angle, elevation, tilt and framing (eye-level, high angle,\n overhead crane, low angle, ...). If it asks for a high or overhead\n angle, draw that angle — the map being top-down does NOT mean the\n frame must be eye-level.\n\nSTYLE — LIGHT STORYBOARD FRAME: thin uniform dark outlines on plain\npaper. No shading, no hatching, no texture, no facial features beyond\na minimal head shape, no costume detail. Human figures are correctly\nproportioned full-body outlines in their canonical poses. Environment\nis reduced to a few placement lines (ground line, key landmark\nsilhouettes) — just enough to read positions and depth.\n\nMAP SLOT MEANINGS:\nE1 = frozen seated man\n\nCAMERA DIRECTION (vertical angle / elevation / framing SOT): Track laterally past the former position of the photograph and settle close at 인우's upper-chest height, holding his frozen upper body in a near-frontal three-quarter composition. His widened eyes sit in the upper central area while the photograph remains just outside his eyeline, and the frame leaves a small amount of space behind him toward the route he will abruptly take.\n\nTHE LOCATION: 소형 어선 갑판\n\nSHOT TEXT (authoritative, Korean): 눈앞의 사진에 시선을 고정한 채 두 눈을 크게 뜨고 얼어붙은 인우의 상체.\n\nCRITICAL — THE MAP IS A STAGING DIAGRAM, NOT CONTENT: do NOT draw the\nmap itself, and do NOT draw any circles, wedges, dots, slot codes\n(E1, E2, CAM) or any other marker symbols in the frame. The output is\na clean storyboard drawing of the scene as seen by the camera only.\n\nNo text, no letters, no numbers, no captions anywhere in the image.", "reused": true, "control_asset_id": "213bc3c3-ec39-417b-be6c-34a30c288724", "asset_id": "657509eb-43ad-415d-af41-d2cdaba44e87" }, "S30sh1": { "lane": "map_marker", "group_id": "bg_small_fishing_boat", "segment_id": "SEG_02", "status": "ok", "image_path": "/Users/manta/Documents/Projects/TheRoad-I1/projects/ecedba66-2474-4dff-a1a4-7d13bac08e1d/images/5ea07552-480a-4fdd-a888-b6147ff239f5/conti/lane_sketch_S30sh1.png", "control_path": "/Users/manta/Documents/Projects/TheRoad-I1/projects/ecedba66-2474-4dff-a1a4-7d13bac08e1d/images/5ea07552-480a-4fdd-a888-b6147ff239f5/conti/lane_control_S30sh1.png", "base_map_path": "/Users/manta/Documents/Projects/TheRoad-I1/projects/ecedba66-2474-4dff-a1a4-7d13bac08e1d/images/background_chain/canon_bg_small_fishing_boat_map.png", "base_map_asset_id": "344cb211-e88c-420d-b314-231478857b60", "seed_path": null, "seed_asset_id": null, "geometry": { "camera": { "look_target": { "x": 0.5, "y": 0.4 }, "origin": { "x": 0.79, "y": 0.89 } }, "entity_placements": [ { "slot": "E1", "subject_en": "moving small fishing boat", "x": 0.5, "y": 0.51 } ], "rationale_ko": "소형 어선은 선체(S1) 중심에 두고 뱃머리(D2)가 화면 상단 중앙을 향하도록 설정했다. 카메라는 선미 뒤쪽 우현 사선의 주변 바다(S2)에 멀리 배치해 낮은 높이의 트레일링 3/4 와이드 구도를 만들며, 시선은 안개에 반쯤 잠기는 전방 갑판과 뱃머리 쪽을 향한다." }, "leakage": { "has_marker_leakage": false, "details_ko": "마커, 웨지, 슬롯 코드 등이 없는 정상적인 스케치입니다." }, "prompt": "Convert the attached TOP-DOWN MARKER MAP into a single CAMERA-VIEW\nstoryboard frame (16:9), seen from the camera marked on the map.\n\nJurisdiction of sources — follow both, they do not conflict:\n- The map is the SOT for XY POSITIONS and horizontal heading ONLY.\n Each red circle labeled E1, E2, ... is where a subject stands; the\n blue dot labeled CAM is the camera position; the blue wedge is its\n horizontal viewing direction and field of view. Subjects outside\n the wedge are out of frame. Distances, directions and relative\n depths between subjects come from the map.\n- The CAMERA DIRECTION text below (when present) is the SOT for the\n VERTICAL angle, elevation, tilt and framing (eye-level, high angle,\n overhead crane, low angle, ...). If it asks for a high or overhead\n angle, draw that angle — the map being top-down does NOT mean the\n frame must be eye-level.\n\nSTYLE — LIGHT STORYBOARD FRAME: thin uniform dark outlines on plain\npaper. No shading, no hatching, no texture, no facial features beyond\na minimal head shape, no costume detail. Human figures are correctly\nproportioned full-body outlines in their canonical poses. Environment\nis reduced to a few placement lines (ground line, key landmark\nsilhouettes) — just enough to read positions and depth.\n\nMAP SLOT MEANINGS:\nE1 = moving small fishing boat\n\nCAMERA DIRECTION (vertical angle / elevation / framing SOT): Near boat height in a trailing three-quarter position, the camera tracks with the small fishing boat as its bow advances toward the upper center and becomes half obscured by the fog while the surrounding sea remains broadly visible. The vessel stays a relatively small silhouette near the middle of the wide frame; compared with the preceding flow stage, only the camera distance closes while the boat's forward axis and sunset illumination remain continuous.\n\nTHE LOCATION: 소형 어선 갑판\n\nSHOT TEXT (authoritative, Korean): 황금빛 노을이 덮인 바다 위, 뱃머리가 거대한 안개벽 속으로 반쯤 가려진 채 물살을 가르며 나아가는 소형 어선의 실루엣.\n\nCRITICAL — THE MAP IS A STAGING DIAGRAM, NOT CONTENT: do NOT draw the\nmap itself, and do NOT draw any circles, wedges, dots, slot codes\n(E1, E2, CAM) or any other marker symbols in the frame. The output is\na clean storyboard drawing of the scene as seen by the camera only.\n\nNo text, no letters, no numbers, no captions anywhere in the image.", "reused": true, "control_asset_id": "884ebab0-10be-4675-83ca-158cac5625ee", "asset_id": "97d17993-0e2c-4372-9b6b-f740fbbfbbbe" } } }, "step_id": "shot_conti_light", "run_id": "645a547c-ca09-43af-b74d-da94715e57ad", "resolved_model": "gpt-image-2", "updated_at": "2026-07-16T12:10:03.297581+00:00", "project_config_snapshot": {} }