{ "defaults": { "duration_seconds": 8.0, "steps": 8 }, "shots": [ { "id": "scene_01", "prompt": "subject_definitions:\n<Subject 1> is a young Japanese anime woman.\n\n<Subject 2> is a male narrator hiding behind the camera.\n\n<Subject 3> is a red one-piece dress.\n\n<Picture 1> is a front-face reference of <Subject 1>.\n\n<Picture 2> is a full-body reference of <Subject 1> standing and peering toward the viewer.\n\n<Picture 4> is a neck-down reference of <Subject 1> in black underwear beneath the red one-piece dress.\n\n<Picture 5> is a face reference of the male narrator <Subject 2>.\n\n<Picture 6> is a reference of <Subject 1> in black underwear.\n\n<Subject 1> is a young Japanese anime woman. <Subject 2> is a male narrator hiding behind the camera. <Subject 3> is a red one-piece dress. <Subject 4> is a black belt. <Subject 5> is a male narrator. <Subject 6> is the woman (<Subject 1>). <Picture 1> is a front-face reference of <Subject 1>. <Picture 2> is a full-body reference of <Subject 1>. <Picture 4> is a neck-down reference of <Subject 1> in black underwear beneath the red one-piece dress. <Picture 5> is a face reference of <Subject 2>. <Picture 6> is a reference of <Subject 1> in black underwear. The setting is a hotel room with a bed. The style is R-15 anime animation.\n\nsummary:\n[reference generation] <Subject 1> sits on a hotel bed wearing a red dress.\n\nretention_analysis:\n<Subject 1> (appears in [Shot 1]): fully_preserved - defined identity and features are retained.\n<Subject 3> (appears in [Shot 1]): fully_preserved - defined identity and features are retained.\n<Subject 4> (appears in [Shot 1]): fully_preserved - defined identity and features are retained.\n\ndetailed_description:\nThe target video continues as a live-action take.\n[Shot 1] <Subject 1> wears <Subject 3> and <Subject 4>.\n<Subject 1> is sitting on a hotel bed, facing the camera with a restless expression. She is wearing the <Subject 3> (<Subject 3>) which is fastened at the waist with a <Subject 4> (<Subject 4>). The camera captures her from a medium shot, showing her slightly fidgety posture. There is absolutely no human speech, voice, vocalization, whispering, muttering, or voiceover in this shot. All characters remain completely silent with their lips closed.\n\noverall_soundscape:\nSoft rustle of fabric against the bedsheets. faint room ambiance. No human voices or vocal sounds.\n\nnon_diegetic_music:\nN/A" }, { "id": "scene_02", "prompt": "subject_definitions:\n<Subject 1> is a young Japanese anime woman.\n\n<Subject 2> is a male narrator hiding behind the camera.\n\n<Subject 3> is a red one-piece dress.\n\n<Picture 1> is a front-face reference of <Subject 1>.\n\n<Picture 2> is a full-body reference of <Subject 1> standing and peering toward the viewer.\n\n<Picture 4> is a neck-down reference of <Subject 1> in black underwear beneath the red one-piece dress.\n\n<Picture 5> is a face reference of the male narrator <Subject 2>.\n\n<Picture 6> is a reference of <Subject 1> in black underwear.\n\n<Subject 1> is a young Japanese anime woman. <Subject 2> is a male narrator hiding behind the camera. <Subject 3> is a red one-piece dress. <Subject 4> is a black belt. <Subject 5> is a male narrator. <Subject 6> is the woman (<Subject 1>). <Picture 1> is a front-face reference of <Subject 1>. <Picture 2> is a full-body reference of <Subject 1>. <Picture 4> is a neck-down reference of <Subject 1> in black underwear beneath the red one-piece dress. <Picture 5> is a face reference of <Subject 2>. <Picture 6> is a reference of <Subject 1> in black underwear. The setting is a hotel room with a bed. The style is R-15 anime animation.\n\nsummary:\n[reference generation] <Subject 1> answers a question about her age and smiles.\n\nretention_analysis:\n<Subject 1> (appears in [Shot 1]): fully_preserved - defined identity and features are retained.\n<Subject 3> (appears in [Shot 1]): fully_preserved - defined identity and features are retained.\n<Subject 4> (appears in [Shot 1]): fully_preserved - defined identity and features are retained.\n\ndetailed_description:\nThe target video continues as a live-action take. The off-screen man says: <d>[Japanese]えーと、年齢は？</d> [Shot 1] <Subject 1> wears <Subject 3> and <Subject 4>. <Subject 1> remains seated on the bed. An adult Japanese male voice is heard from off-screen. <Subject 1> remains silent. Then, <Subject 1> (S1) says <d>[Japanese]…２１です</d> She looks toward the camera and gives a small smile. The camera maintains a steady medium shot. Only the written <d> lines are spoken aloud; do not add any other dialogue, mutter, whisper, or lead-in voiceover. Spoken <d> lines are Japanese only.\n\noverall_soundscape:\nSoft rustle of fabric, slight movement of her body.\n\nnon_diegetic_music:\nN/A" }, { "id": "scene_03", "prompt": "subject_definitions:\n<Subject 1> is a young Japanese anime woman.\n\n<Subject 2> is a male narrator hiding behind the camera.\n\n<Subject 3> is a red one-piece dress.\n\n<Picture 1> is a front-face reference of <Subject 1>.\n\n<Picture 2> is a full-body reference of <Subject 1> standing and peering toward the viewer.\n\n<Picture 4> is a neck-down reference of <Subject 1> in black underwear beneath the red one-piece dress.\n\n<Picture 5> is a face reference of the male narrator <Subject 2>.\n\n<Picture 6> is a reference of <Subject 1> in black underwear.\n\n<Subject 1> is a young Japanese anime woman. <Subject 2> is a male narrator hiding behind the camera. <Subject 3> is a red one-piece dress. <Subject 4> is a black belt. <Subject 5> is a male narrator. <Subject 6> is the woman (<Subject 1>). <Picture 1> is a front-face reference of <Subject 1>. <Picture 2> is a full-body reference of <Subject 1>. <Picture 4> is a neck-down reference of <Subject 1> in black underwear beneath the red one-piece dress. <Picture 5> is a face reference of <Subject 2>. <Picture 6> is a reference of <Subject 1> in black underwear. The setting is a hotel room with a bed. The style is R-15 anime animation.\n\nsummary:\n[reference generation] <Subject 1> reacts to a compliment about her figure.\n\nretention_analysis:\n<Subject 1> (appears in [Shot 1]): fully_preserved - defined identity and features are retained.\n<Subject 3> (appears in [Shot 1]): fully_preserved - defined identity and features are retained.\n<Subject 4> (appears in [Shot 1]): fully_preserved - defined identity and features are retained.\n\ndetailed_description:\nThe target video continues as a live-action take. The off-screen man says: <d>[Japanese]いいスタイルしてますねー</d> [Shot 1] <Subject 1> wears <Subject 3> and <Subject 4>. <Subject 1> is still on the bed. An adult Japanese male voice is heard from off-screen. <Subject 1> remains silent. Then, <Subject 1> (S1) says <d>[Japanese]いえ、そんなことは…</d> Her face flushes slightly red as she shakes her head in denial, looking at the camera. Only the written <d> lines are spoken aloud; do not add any other dialogue, mutter, whisper, or lead-in voiceover. Spoken <d> lines are Japanese only.\n\noverall_soundscape:\nSoft rustle of fabric, slight movement of her head.\n\nnon_diegetic_music:\nN/A" }, { "id": "scene_04", "prompt": "subject_definitions:\n<Subject 1> is a young Japanese anime woman.\n\n<Subject 2> is a male narrator hiding behind the camera.\n\n<Subject 3> is a red one-piece dress.\n\n<Picture 1> is a front-face reference of <Subject 1>.\n\n<Picture 2> is a full-body reference of <Subject 1> standing and peering toward the viewer.\n\n<Picture 4> is a neck-down reference of <Subject 1> in black underwear beneath the red one-piece dress.\n\n<Picture 5> is a face reference of the male narrator <Subject 2>.\n\n<Picture 6> is a reference of <Subject 1> in black underwear.\n\n<Subject 1> is a young Japanese anime woman. <Subject 2> is a male narrator hiding behind the camera. <Subject 3> is a red one-piece dress. <Subject 4> is a black belt. <Subject 5> is a male narrator. <Subject 6> is the woman (<Subject 1>). <Picture 1> is a front-face reference of <Subject 1>. <Picture 2> is a full-body reference of <Subject 1>. <Picture 4> is a neck-down reference of <Subject 1> in black underwear beneath the red one-piece dress. <Picture 5> is a face reference of <Subject 2>. <Picture 6> is a reference of <Subject 1> in black underwear. The setting is a hotel room with a bed. The style is R-15 anime animation.\n\nsummary:\n[reference generation] <Subject 1> removes her belt and stands up.\n\nretention_analysis:\n<Subject 1> (appears in [Shot 1]): fully_preserved - defined identity and features are retained.\n<Subject 3> (appears in [Shot 1]): fully_preserved - defined identity and features are retained.\n<Subject 4> (appears in [Shot 1]): fully_preserved - defined identity and features are retained.\n\ndetailed_description:\nThe target video continues as a live-action take. The off-screen man says: <d>[Japanese]じゃあ、脱いでもらえる？</d> [Shot 1] <Subject 1> wears <Subject 3> and <Subject 4>. An adult Japanese male voice is heard from off-screen. <Subject 1> smiles shyly. She reaches for the <Subject 4> (<Subject 4>) at her waist, unfastens it, and takes it off, placing it on the floor in front of the bed. She then stands up from the bed. Only the written <d> line is spoken aloud. It is spoken exclusively by the off-screen adult male speaker in a clearly male Japanese voice. The female character remains silent throughout the entire shot. Do not use the female character's voice for the <d> line. Do not add any other dialogue, mutter, whisper, or lead-in voiceover. Spoken dialogue is Japanese only.\n\noverall_soundscape:\nSound of the belt buckle clicking, fabric rustling, feet hitting the floor.\n\nnon_diegetic_music:\nN/A" }, { "id": "scene_05", "prompt": "subject_definitions:\n<Subject 1> is a young Japanese anime woman.\n\n<Subject 2> is a male narrator hiding behind the camera.\n\n<Subject 3> is a red one-piece dress.\n\n<Picture 1> is a front-face reference of <Subject 1>.\n\n<Picture 2> is a full-body reference of <Subject 1> standing and peering toward the viewer.\n\n<Picture 4> is a neck-down reference of <Subject 1> in black underwear beneath the red one-piece dress.\n\n<Picture 5> is a face reference of the male narrator <Subject 2>.\n\n<Picture 6> is a reference of <Subject 1> in black underwear.\n\n<Subject 1> is a young Japanese anime woman. <Subject 2> is a male narrator hiding behind the camera. <Subject 3> is a red one-piece dress. <Subject 4> is a black belt. <Subject 5> is a male narrator. <Subject 6> is the woman (<Subject 1>). <Picture 1> is a front-face reference of <Subject 1>. <Picture 2> is a full-body reference of <Subject 1>. <Picture 4> is a neck-down reference of <Subject 1> in black underwear beneath the red one-piece dress. <Picture 5> is a face reference of <Subject 2>. <Picture 6> is a reference of <Subject 1> in black underwear. The setting is a hotel room with a bed. The style is R-15 anime animation.\n\nsummary:\n[reference generation] <Subject 1> removes her dress to reveal her black underwear.\n\nretention_analysis:\n<Subject 1> (appears in [Shot 1]): fully_preserved - defined identity and features are retained.\n<Subject 3> (appears in [Shot 1]): fully_preserved - defined identity and features are retained.\n<Subject 4> (appears in [Shot 1]): fully_preserved - defined identity and features are retained.\n\ndetailed_description:\nThe target video continues as a live-action take.\n[Shot 1] <Subject 1> wears <Subject 3>. <Subject 1> holds <Subject 4>.\n<Subject 1> slowly slides her arms out of the sleeves of the <Subject 3> (<Subject 3>) one by one. She holds the <Subject 3> with both hands, revealing the black bra underneath. She then releases the <Subject 3> and lets it drop to the floor at her feet, revealing the black panties as well. She is now standing in only the black bra and black panties. She smiles awkwardly at the camera. There is absolutely no human speech, voice, vocalization, whispering, muttering, or voiceover in this shot. All characters remain completely silent with their lips closed.\n\noverall_soundscape:\nFabric sliding over skin. the heavy thud of the dress hitting the floor. No human voices or vocal sounds.\n\nnon_diegetic_music:\nN/A" }, { "id": "scene_06", "prompt": "subject_definitions:\n<Subject 1> is a young Japanese anime woman.\n\n<Subject 2> is a male narrator hiding behind the camera.\n\n<Subject 3> is a red one-piece dress.\n\n<Picture 1> is a front-face reference of <Subject 1>.\n\n<Picture 2> is a full-body reference of <Subject 1> standing and peering toward the viewer.\n\n<Picture 4> is a neck-down reference of <Subject 1> in black underwear beneath the red one-piece dress.\n\n<Picture 5> is a face reference of the male narrator <Subject 2>.\n\n<Picture 6> is a reference of <Subject 1> in black underwear.\n\n<Subject 1> is a young Japanese anime woman. <Subject 2> is a male narrator hiding behind the camera. <Subject 3> is a red one-piece dress. <Subject 4> is a black belt. <Subject 5> is a male narrator. <Subject 6> is the woman (<Subject 1>). <Picture 1> is a front-face reference of <Subject 1>. <Picture 2> is a full-body reference of <Subject 1>. <Picture 4> is a neck-down reference of <Subject 1> in black underwear beneath the red one-piece dress. <Picture 5> is a face reference of <Subject 2>. <Picture 6> is a reference of <Subject 1> in black underwear. The setting is a hotel room with a bed. The style is R-15 anime animation.\n\nsummary:\n[reference generation] <Subject 1> answers a question about her cup size while sitting back down.\n\nretention_analysis:\n<Subject 7> (appears in [Shot 1]): fully_preserved - defined identity and features are retained.\n<Subject 8> (appears in [Shot 1]): fully_preserved - defined identity and features are retained.\n<Subject 1> (appears in [Shot 1]): fully_preserved - defined identity and features are retained.\n<Subject 4> (appears in [Shot 1]): fully_preserved - defined identity and features are retained.\n<Subject 3> (appears in [Shot 1]): fully_preserved - defined identity and features are retained.\n\ndetailed_description:\nThe target video continues as a live-action take. The off-screen man says: <d>[Japanese]結構大きいね！何カップ？</d> [Shot 1] <Subject 7> is the panties. <Subject 8> is the bra. <Subject 1> wears <Subject 7> and <Subject 8>. <Subject 1> holds <Subject 4>. <Subject 3> is off, on the floor. The camera zooms in on <Subject 1>'s chest, highlighting the black bra. An adult Japanese male voice is heard from off-screen. <Subject 1>'s eyes wander upward, looking slightly away from the camera. <Subject 1> (S1) says <d>[Japanese]Ｆ…くらいかな？</d> She gives a polite smile and sits back down on the bed in her black underwear. Only the written <d> lines are spoken aloud; do not add any other dialogue, mutter, whisper, or lead-in voiceover. Spoken <d> lines are Japanese only.\n\noverall_soundscape:\nSoft rustle of the bra fabric, movement of her body sitting down.\n\nnon_diegetic_music:\nN/A" }, ～～～～～～～～