Image to video

FORMAT 15 seconds | 16:9 | photoreal live-action war thriller Fictional Sahelian city at blue-hour dawn. Urgent, human, suspenseful, non-graphic. REFERENCE CONTROL Use the supplied first frame to lock AMINA’s identity and wardrobe. Preserve face, close-cropped hair, deep-brown skin, navy field jacket, silver earpiece, canvas satchel, analogue radio, boots and determined expression. Build the surrounding environment from this description, without requiring a second image: A concrete bridge, dry riverbed, dense low-rise city, rooftop antennae, dusty atmosphere, cool-blue shadows and restrained amber dawn. No real countries, flags, organizations or landmarks. STORY Amina is the last radio operator keeping an evacuation bridge open. Communications fail during a distant attack. She must activate the backup relay and transmit clearance before automatic controls close the bridge. 00–02s — CHANNEL OFFLINE Begin with Amina on the rooftop in the supplied first frame, then cut to a close-up of her radio. Amina grips the analogue radio as red emergency lights pulse. Dust trembles, cables quiver, broken static fills the room. Automated voice: “Emergency channel offline.” She looks up and immediately moves. 02–04s — BRIDGE CLOSING Wide aerial-oblique rooftop view. Civilians cross the bridge through windblown dust. Checkpoint signal flickers green → red as shutdown begins. A distant shockwave moves through city haze. No visible casualties. 04–06.5s — RUN Low handheld tracking behind Amina running through a damaged service corridor. Emergency lights flash, dust falls, cables briefly spark, boots hit wet concrete. She pushes through a heavy metal door into blowing dust. 06.5–09s — FLASHBACK Two minutes earlier. Medium side-profile on rooftop. Amina sees the unstable bridge signal, manually rotates the antenna and connects her radio to backup power. Slow camera push. Wind moves her satchel strap as first amber sunlight reaches her jacket. 09–12.5s — TRANSMISSION Rapid cross-cut montage: frequency dial locks → antenna aligns → bridge cables vibrate → red signal flashes → Amina speaks into radio: “Channel Seven, keep the bridge open.” Overhead shot: civilians continue crossing. Sharp hard cuts, brief black frames, natural motion blur. 12.5–15s — SIGNAL RESTORED Sound suddenly drops. Amina listens on the rooftop. Clear radio tone. Checkpoint signal changes red → steady green. Evacuation continues. Amina exhales and watches the bridge as dawn reaches the city. Slow push toward her face. Cut to black. CAMERA Large-format live-action realism. Precise nonlinear cross-cutting. Handheld tracking, aerial-oblique coverage, restrained push-ins and close details. Natural motion blur, fine film grain, subtle gate weave. No impossible camera movement. ENVIRONMENT + PHYSICS Cool-blue shadows against muted amber sunrise. Deep blacks, dusty atmosphere, weathered concrete. Wind realistically affects dust, clothing, cables and antenna equipment. All movement and interactions remain physically believable. AUDIO Stereo radio static, electrical hum, wind, boots, antenna creaks, distant low artillery and helicopter rotors. Restrained suspenseful sub-bass builds through the montage. VISUAL Photoreal cinematic war-thriller footage. Immense scale contrasted with one determined operator. Grounded, tense and human. No subtitles, text, logos, watermarks, political messaging, flags or military insignia. CONTINUITY Same Amina identity, wardrobe, radio and satchel throughout. Same bridge, city, atmosphere and dawn lighting. Maintain geographic and temporal continuity across all shots.

{ "cinematic_video_request": { "meta": { "title": "Sadie Sink - Night Drive Portrait", "style_preset": "Cinematic Realism", "duration_seconds": 10, "resolution": "4K", "aspect_ratio": "9:16" }, "prompts": { "main_prompt": "Ultra-realistic cinematic animation of Sadie Sink sitting in the backseat of a luxury car at night. The city lights outside the window create soft motion blur with passing traffic and glowing streetlights. Subtle camera push-in shot from medium frame to close-up. Gentle movement in her hair as the car moves. She slowly shifts her gaze toward the window, blinking naturally, then looks back toward the camera with a calm, slightly mysterious expression.", "visual_modifiers": "4K cinematic quality, realistic skin texture, natural facial micro-expressions, smooth motion, dramatic nighttime mood, film-grade color grading, soft contrast, subtle handheld camera feel, shallow depth of field, bokeh city lights, detailed leather texture.", "lighting_prompt": "Soft ambient lighting from streetlights flickers across her face, creating dynamic shadows and warm highlights. Background traffic lights streak smoothly past the window." }, "scene_specifications": { "subject": { "name": "Sadie Sink", "action": "Sitting, gazing out window, turning head to camera, blinking", "expression": "Calm, mysterious, natural micro-expressions", "details": "Gentle hair movement, realistic skin texture" }, "environment": { "setting": "Backseat of luxury car", "time": "Night", "exterior": "City streets, passing traffic", "details": "Detailed leather interior, window reflections" }, "camera": { "movement": "Slow push-in (dolly forward)", "stabilization": "Slight handheld feel (organic motion)", "framing": "Medium shot to Close-up", "focus": "Shallow depth of field with background bokeh" }, "lighting": { "type": "Dynamic/Transient", "sources": "Passing streetlights, city glow", "characteristics": "Warm highlights, soft contrast, rhythmic flickering shadows" } } } }

Use the uploaded image as the exact visual reference. Preserve the same young woman’s face, pastel pink twin pigtails, oversized faded red vintage T-shirt with yellow lettering, night street environment, wet reflective ground, harsh direct flash photography look, dark urban background, raw Y2K anime inspired street fashion mood. Keep the identity, outfit, pose language, lighting, and gritty night atmosphere consistent. A cinematic music video style night street portrait begins from the reference frame. The camera holds a medium close up with slight handheld movement, as if shot with an on camera flash. Rain wet pavement glows behind her with blurred city lights and reflections. The woman gently adjusts both pink pigtails with her hands, then slowly turns her eyes toward the camera with a cool, rebellious expression. Her lips part naturally as she speaks, with subtle realistic lip sync. Her hair moves slightly in the night breeze. The flash creates bright highlights on her skin and hair while the background stays dark and gritty.[Character A: Pink-haired street fashion girl, soft confident young female voice, slightly playful, calm attitude]: “You thought I came here to disappear?”She pauses, tilts her head a little, still holding one pigtail. The camera slowly pushes in closer, keeping the raw flash-photo texture and shallow depth of field.[Character A: same voice, quieter but more confident]: “No… tonight, I become impossible to ignore.”Ambient sound: distant city traffic, soft rain dripping, faint underground music beat in the background. No captions, no subtitles, no extra text on screen. Keep the motion realistic, stylish, and cinematic. Avoid face distortion, avoid changing the outfit, avoid changing hair color, avoid adding new characters.

She begins standing still, facing camera with a calm confident expression. Slowly she raises one hand to lift her hair off her neck, then turns away from the camera revealing "RAPHINHA 11" printed in gold on the back of her Barcelona jersey. She gazes out over the stadium pitch. Her hair lifts and settles naturally in a light wind. The giant Barcelona tifo mosaic on the pitch fills the background. Camera holds a steady handheld medium shot with subtle organic sway. Overcast daylight, cool diffused light, cinematic 4K, realistic hair movement, authentic pre-match stadium energy.

风格：真实地铁车厢偷拍感，轻微手持晃动，车厢自然冷白灯，轻微压缩画质，微弱运动模糊。 0-3 秒 女生坐在金属座椅上，身体靠着车厢墙，低头看手机。她一只手拿手机，另一只手自然搭在棕色小包附近，白色耳机线垂在胸前和大腿上。表情放松、没有防备。镜头从斜前方拍摄，左侧有模糊立柱遮挡，地铁轻微晃动。 3-5 秒 她像是察觉有人在拍，手指停顿一下，眼神先从手机屏幕上移到镜头方向。不要突然抬头，要慢慢抬眼：先眼睛动，再微微抬下巴。表情从放松变成冷静审视，带一点“你在拍我？”的疑惑和压迫感。 5-7 秒 她保持看镜头，嘴唇轻轻抿住，眉眼微微收紧，不笑，不说话。手机还握在手里，身体基本不动，只是肩膀和呼吸有细微起伏。这个阶段重点是眼神：平静、淡定、带一点挑衅和不躲闪。 7-10 秒 她视线仍然锁住镜头，另一只手缓慢移到大腿处，从灰色短裙/裤脚边缘轻轻往上拉一下，短暂露出一点黑色蕾丝边。动作幅度很小、真实、自然，不夸张，不大幅裸露。拉完后手停在大腿旁边，她仍然看着镜头，表情很淡，像是在无声回应拍摄者。 声音 无对白。只有地铁行驶低频声、车厢轻微震动、远处模糊乘客声。 负面提示词 不要字幕，不要水印，不要换场景，不要站起来，不要大幅动作，不要夸张微笑，不要说话口型，不要色情化，不要过度裸露。必须是从大腿裤脚/裙摆边缘轻轻拉起，露出一点黑色蕾丝边；不要拉腰部。保持首帧人物、衣服、包、手机、耳机线和地铁结构完全一致。

Animate the supplied courtside image as a live basketball-broadcast audience cutaway. Preserve the woman’s face, hair, clothing, seated posture, nearby spectators and arena lighting. She notices the camera, blinks naturally, turns her head slightly and gives a relaxed smile. Use a steady broadcast-style medium shot with a subtle optical push-in, realistic telephoto depth, natural skin and slight broadcast compression. Background spectators move gently without changing identity or seats. Add low crowd ambience and distant court sounds, no dialogue. Do not introduce new captions, logos, extra people, distorted hands or a different venue.

Use the uploaded image as the first frame of a tense, realistic exchange in a worn boxing gym. Preserve the two adult men, their faces, clothes, ring, hanging punching bags and warm overhead lights. The young boxer on the left speaks emphatically, gesturing with both hands and leaning toward the older trainer. The trainer initially keeps his arms folded and holds a stern, patient expression, then answers quietly. Boxer: "I am not giving up." Trainer: "Then keep your guard up." Natural English lip sync, distinct voices, short pauses, visible breathing and restrained facial reactions. Begin with the two-shot, cut closer to the boxer, then to the trainer’s response. Keep screen direction and eye contact consistent. Low gym room tone, faint equipment sounds; no music, subtitles, face changes, duplicated hands or physical fighting.

Duration: 20秒 Aspect ratio: 9:16 Overall style: POV男友第一人称短剧，真实手机手持轻微呼吸晃动，室内自然光通透，手机滑动+点击瞬间换装，节奏轻快甜宠，竖屏 【场景】明亮客厅/卧室，落地窗自然光，背景简洁温馨，地面干净无杂物 【角色】女主（所提供的首帧图，开场穿角色原本服装）；男主为POV视角，全程不露脸，仅画面下方出现持手机的男性手 【道具】一部现代全面屏手机，竖持。屏幕是简洁时尚穿搭App：上方大预览图，下方横向可滑动的服装卡片。每次滑动有跟手惯性，点击卡片时屏幕轻微缩放+短震反馈 SHOT 1（00:00-00:03.5）开场展示 画面：POV，女主站画面中央，开心地原地转半圈展示穿搭，双手轻提衣摆，眼睛亮亮看镜头。 台词：女主："老公，看我今天穿得怎么样？" 音效：室内底噪，衣料轻响。 Constraints: 女主全身入画，开场服装=角色原本服装。 SHOT 2（00:03.5-00:06.5）掏出手机+第一次滑动 画面：画面下方男主手举起手机入镜，屏幕亮起，显示穿搭App。拇指从右向左滑过卡片，停在白衬衫配蓝色牛仔裤的卡片上，预览图放大。 台词：男主（画外音，慵懒）："还行。换这套看看。" 音效：屏幕滑动轻摩擦声。 Constraints: 手机屏幕清晰可读，卡片上能看出是白衬衫配蓝色牛仔裤；男主只出手，不露脸。 SHOT 3（00:06.5-00:09.5）第一次点击瞬变 画面：拇指点在白衬衫配蓝色牛仔裤卡片上，屏幕短震。点击落下的同一帧，女主服装瞬间变成白衬衫配蓝色牛仔裤，站位、姿势、发型完全连续；她低头看衣服，瞪大眼张嘴，再抬头看镜头。 台词：女主（惊讶）："诶？！" 音效：清脆"滴"+换装"叮"，衣料展开微响。 Constraints: 换装发生在点击同一帧，单帧瞬变；可加一道从上到下扫过身体的柔和微光；禁止叠化、黑闪、人物跳位。 SHOT 4（00:09.5-00:13）第二次滑动+点击瞬变 画面：拇指再次左滑，卡片停在浅蓝色连衣裙，点一下。同一帧女主换成浅蓝色连衣裙；这次她反应过来，脸颊微红，手指绞在身前，害羞小幅扭身，眼神躲闪又忍不住看镜头。 台词：男主（画外音）："这套也行。" 女主（小声）："你……乱点。" 音效：滑动声+"滴"+"叮"，轻快音乐渐起。 Constraints: 滑动必须完整看见从白衬衫配蓝色牛仔裤卡片滑到浅蓝色连衣裙；换装仍是点击同一帧硬切。 SHOT 5（00:13-00:16.5）第三次滑动+点击瞬变 画面：再滑到针织开衫配百褶裙，点一下。同一帧换成针织开衫配百褶裙；女主已经习惯，咬一下下唇，抬手整理新衣服领口/裙摆，眼神带点得意看镜头。 台词：女主："这套……还挺好看的嘛。" 音效："滴"+"叮"，音乐节奏加快半拍。 Constraints: 三次反应递进：惊讶→害羞→得意；每次都有小动作，禁止呆立。 SHOT 6（00:16.5-00:20）第四次滑动+点击+收尾 画面：最后一滑停在米色外套配长裤，拇指点下。同一帧女主瞬变为米色外套配长裤；她直接笑出来，上前一步凑近镜头，狡黠眯眼。手机仍举在画面下方，屏幕停在米色外套配长裤预览。 台词：男主（画外音，满意）："这套定了。" 女主（笑着凑近，嗔怪）："你好坏啊——我好喜欢！" 音效："滴"+"叮"，音乐在女主台词落点收在甜甜重拍。 Constraints: 结尾女主凑近但不过分贴脸；手机始终在画面下方可见；服装顺序锁定：开场=原本 → 白衬衫配蓝色牛仔裤 → 浅蓝色连衣裙 → 针织开衫配百褶裙 → 米色外套配长裤，不可乱。 【导演铁律】 1. 换装铁律：每次换装发生在“点击”同一帧，单帧瞬变。女主站位、姿势、发型、脸型跨越换装帧完全连续，只有服装改变。 2. POV铁律：镜头=男主眼睛，全程轻微手持呼吸晃动；男主只出手+画外音，禁止露脸。 3. 手机铁律：每次必须先滑动再点击，滑动跟手、卡片切换清楚；禁止凭空换装、禁止不看屏幕直接变。 4. 服装引用：开场=所提供的首帧图原本衣服；第1次=白衬衫配蓝色牛仔裤；第2次=浅蓝色连衣裙；第3次=针织开衫配百褶裙；第4次=米色外套配长裤。 5. 台词口型清晰，与点击节拍对齐。 Negative: 叠化换装，黑闪，白闪，烟雾，旋转遮挡，人物瞬移，姿势断裂，脸型变化，发型变化，鞋子乱变，露男主脸，横屏，字幕，水印，Logo，背景突然更换，多人入画，镜头剧烈抖动，手机屏幕模糊不可读。

A cinematic 30-second alpine winter vlog montage featuring the same young woman exploring a snow-covered mountain village during a cozy winter getaway. Shot like an authentic luxury travel diary with handheld camera movement, candid moments, soft overcast daylight transitioning to warm firelight, dreamy 35mm film aesthetics, cool-toned color grading with warm amber accents, shallow depth of field, natural skin texture, atmospheric lighting, and cinematic storytelling. Maintain the same woman throughout every scene: dark hair, youthful appearance, natural makeup, relaxed happy expression — identity locked to the reference image. Format: 4K cinematic video, 24fps, 35mm film grain, realistic handheld camera, soft focus, cool-warm contrast palette, travel documentary style. Scene 1 (0-4s) — Arrival at the Chalet Village: A crisp winter morning, light snow falling. She steps out of a wooden chalet doorway wearing a cream cable-knit sweater, faux-fur-trimmed parka, and a knit beanie, breath visible in the cold air. Camera trails her from behind down a snowy cobblestone lane lined with timber cottages, then swings into a close-up as she turns and smiles, saying softly, "Okay, this already feels like a movie." Scene 2 (4-8s) — First Ski Run: She glides down a gentle beginner slope, snow spraying softly at her skis, pine trees blurred past on either side. Low chase shots from behind, then a front-facing GoPro-style angle catching her laughing as she wobbles slightly and steadies herself. Scene 3 (8-12s) — Snow-Covered Pine Forest Walk: A worm's-eye view looking up through frosted pine branches, pale sunlight filtering through with soft lens flares on ice crystals. Cut to a close-up of her walking through untouched snow, mittened hand trailing along a low branch, murmuring, "It's so quiet here it feels unreal." Scene 4 (12-16s) — Cozy Café Break: She sits by a frosted window inside a small timber café, wrapped hands around a steaming mug of hot chocolate topped with whipped cream. Warm interior light contrasts the blue-white world outside. Close-up on the marshmallows melting, her contented half-smile as she watches snow drift past the glass. Scene 5 (16-20s) — Sledding Adventure: She sits on a wooden sled at the top of a gentle hill, pushes off, and speeds down laughing, hair flying loose from under her beanie. Camera tracks alongside at sled height. She tumbles softly into a snowbank at the bottom, laughing, and calls toward the camera, "Don't film this part — okay, keep filming it." Scene 6 (20-24s) — Christmas Market Exploration: A glowing evening market strung with warm fairy lights, snow settling on wooden stalls, the smell of roasted chestnuts implied through steam rising from vendor carts. She weaves through the crowd sampling mulled cider and gingerbread, face lit by string lights and lantern glow. Cinematic close-ups of her eyes widening at the taste, lights blurred into soft bokeh behind her. Scene 7 (24-27s) — Golden Sunset Over the Peaks: A wide silhouette shot of her standing on a snowy overlook as the sun dips behind the mountain range, sky shifting from pale gold to soft violet, her scarf drifting in the wind. She tilts her face toward the fading light, one hand holding her beanie in place. Scene 8 (27-30s) — Fireside Night Reflection: A cabin interior at night, fireplace glowing, snow visible through a frosted window behind her. She sits curled up on a wool blanket, mug in hand, firelight flickering across her face. Final intimate close-up as she looks directly into the lens with a soft, warm smile, saying, "Goodnight from the mountains." Camera Style: Authentic travel vlog cinematography, handheld camera shake, smooth cinematic transitions, slow push-ins, natural drifting movement, occasional POV shots, realistic autofocus hunting, subtle motion blur. Visual Style: Cozy alpine winter film, boutique travel-brand aesthetic, soft diffused daylight with warm firelight interiors, realistic skin texture, cinematic shallow depth of field, nostalgic 35mm film look, cool-warm contrast tones, natural unscripted expressions, emotional storytelling. Avoid: cartoon style, CGI look, plastic skin, unrealistic face, inconsistent character appearance, changing hairstyle, extra fingers, distorted body, artificial lighting, oversaturated colors, blurry face, unnatural movements, duplicate people.

时长：20秒 画幅：9:16竖屏 主题：同一句韩语，韩剧里与现实中的反差。 形式：单人双状态，五组A/B硬切对照。 【人物绑定】 所提供的首帧图是唯一人物参考。全片保持同一人物身份，面部、发型和服装沿用参考图，不重新设计长相。 A与B是同一个人的两种表演，不是两个人对话，不出现双人同框或分屏。 【状态A｜韩式爱情电影】 韩式爱情电影中的人物近景，精致、温柔、含蓄，带轻微浪漫感。 光线： 窗侧大面积柔光照亮面部，明暗过渡细腻；侧后方一束柔暖逆光轻轻勾出发丝和肩部轮廓。眼睛有自然眼神光，面部立体，不使用均匀磨平五官的美颜光。 调色： 奶油白高光、低饱和暖肤色、轻微青灰冷调阴影。整体清透，黑位略柔，亮部平滑过渡。背景偏米白、淡灰与浅暖色，避免整张画面泛黄、泛粉。 摄影质感： 中长焦人像镜头观感，浅景深，眼睛与嘴部清晰，窗帘和远处小灯柔化为细腻散景。高光带极轻微柔晕，保留皮肤纹理与细微胶片颗粒。 每个A镜头只有极轻、几乎察觉不到的缓慢推近。 表演： 目光柔软，呼吸自然，动作幅度小。开口前有短暂情绪停顿，声音轻柔，语尾轻落，像韩剧中对亲近的人说话。 【状态B｜普通手机直拍】 现实室内手机前置直拍，简单灰墙背景，较直接的正面照明。白平衡中性，反差和锐度比A更直接，背景也更清楚。 取消A的逆光轮廓、浅景深散景、柔晕与电影调色，但仍保持正常曝光和真实肤质。 机位固定，构图比A略紧，保留轻微手机广角近距离感，不夸张拉伸面部。 表演： 直视镜头，眉眼明显发力，身体稍向前探，语气短促、直接、不耐烦。喜剧来自突然变脸和说话方式，不演成失控尖叫。 【构图与剪辑】 统一头顶至腰部的竖屏中近景，眼睛位置保持在画面相近高度。 A说完后直接硬切B，画面质感、神态与声音同时改变。 每次切换都保持同一张脸可辨认，不用变形、闪白、黑帧或变装转场。 SHOT 1｜00–04秒｜别看 Subject： 同一人物，温柔含蓄与直接斥责形成反差。 Action / Performance： 00–02秒，状态A： 人物先看向侧下方，轻轻转眼看向镜头，带一点含蓄笑意，柔声说“别看”。说完略收回目光，保持自然羞涩。 镜头极缓慢推近，柔暖逆光照亮少量发丝。 02–04秒，硬切状态B： 人物直接盯住镜头，眉头收紧，头稍向前探，用明显更重的语气重复“别看”，随即补上粗口。 最后一个字落下，目光仍停在镜头上，留一点不悦的静止。 Dialogue： A：“보지 마.” B：“보지 마, 이 개새끼야.” Environment / Camera / Style： A为柔光窗帘与奶油色散景，韩式爱情电影质感。 B为灰墙、直接光与固定手机近景。 Constraints： 两句均由画面内人物本人说出，先说完A，再切B，口型准确。 SHOT 2｜04–08秒｜别这样 Subject： 同一人物，亲昵劝阻与不耐烦训斥。 Action / Performance： 04–06秒，状态A： 人物轻轻歪头，肩膀放松，目光带一点柔软的责怪，低声说“别这样”。语尾很轻，像面对亲近的人。 面部清楚，背景柔化，镜头保持细微推近。 06–08秒，硬切状态B： 头摆正，目光直直看向镜头，短促地说“别这样”，紧接着补上粗口。 下巴只随重音向前点一下，说完停止动作。 Dialogue： A：“하지 마.” B：“하지 마, 이 개새끼야.” Environment / Camera / Style： 沿用A的精致电影光线与B的普通手机直拍，硬切瞬间清楚显出质感差异。 Constraints： 不用夸张摇头或甩头。表情变化集中在眼神、眉间、下颌与语气。 SHOT 3｜08–12秒｜别走 Subject： 同一人物，真诚挽留与强硬命令。 Action / Performance： 08–10秒，状态A： 人物目光稍微抬起，像在看即将离开的人，嘴唇轻启，用带一点不舍的声音说“别走”。 说完保持期待的眼神，轻微吸气，不落泪、不追出画面。 柔光留住眼睛里的细小反光，镜头缓慢靠近一点。 10–12秒，硬切状态B： 人物突然睁大眼睛，头略向前探，把同一句话说成强硬制止，紧接着补上粗口。 表情像在命令对方站住，保持眼神压迫感。 Dialogue： A：“가지 마.” B：“가지 마, 이 개새끼야.” Environment / Camera / Style： A浅景深、柔和高光、冷暖细腻分离；B中性颜色、清晰背景、固定机位。 Constraints： A的情绪含蓄，B的语气强硬。不要两段都演成哭泣或大喊。 SHOT 4｜12–16秒｜别听 Subject： 同一人物，温柔安慰与暴躁纠正。 Action / Performance： 12–14秒，状态A： 人物轻轻叹气，短暂垂眼，再抬眼看镜头，小幅摆一下手。 先自然说出感叹词，再温柔地说“别听”，像在劝对方不要把难听的话放在心上。 表情有理解与安慰，动作细小。 14–16秒，硬切状态B： 人物眼睛睁大，眉毛抬起，带着明显不耐烦说“别听”，随后快速补上粗口。 头部只做一次短促前倾，咬字清楚、重音明确。 Dialogue： A：“아이고… 듣지 마.” B：“듣지 마, 이 개새끼야.” Environment / Camera / Style： A为韩剧中安慰人的电影近景；B像手机镜头前直接训人。 A的轻柔叹气与B的干脆发声形成声音反差。 Constraints： A的感叹词与后半句自然连贯。两段声音不重叠，不增加其他人的回应。 SHOT 5｜16–20秒｜别哭 Subject： 同一人物，心疼安抚与最后的不耐烦吐槽。 Action / Performance： 16–17秒，状态A： 人物微微向前倾，双手在腰胸之间轻抬，做一个很小的安抚动作，眼神心疼，柔声说“别哭”。 保持电影柔光与清楚的嘴部细节。 17–20秒，硬切状态B： 人物立刻以烦躁口吻说“别哭”，补上粗口。 接着视线移到旁边，肩膀松下来，压低声音快速嘟囔，像已经受够对方总哭。 最后以未说完的“씨…”收尾，闭嘴，留下一个无语的表情。不要突然笑场。 Dialogue： A：“울지 마.” B：“울지 마, 이 개새끼야. 맨날 쳐 울고 앉아있어, 씨…” Environment / Camera / Style： B状态持续到片尾，不再切回A，不加通关式音效或结束标题。 Constraints： 末段先强硬制止，再低声抱怨，音量有变化。尾句语速加快但发音清楚，不扩写新台词。 【声音与口型】 所有韩语都是画面内人物的现场对白，绝不是旁白。 人物说话时，嘴唇、下颌、呼吸与韩语发音同步。 同一基础声线： A轻柔、亲近、带自然气息，不能用气声糊掉发音； B更直接、更响、重音清楚，带不耐烦，不能全程破音尖叫。 保留轻微呼吸、叹气与室内底噪。 A段可有极低音量、稀疏的浪漫钢琴底音；切到B时钢琴立即停止，只剩干净直接的人声与房间底噪。下一段A再轻轻恢复同一音乐动机。 音乐始终让位于台词，不加入罐头笑声。 【核心要求】 一张人物卡，全片同一个人。 五组顺序固定： 보지 마 → 하지 마 → 가지 마 → 듣지 마 → 울지 마。 每组A轻柔说短句，B硬切后重复同一句，再补粗口。 A必须明确呈现韩式爱情电影的布光、调色、浅景深与细腻表演。 B必须保留普通手机直拍的直接感，让电影滤镜的消失成为笑点的一部分。 无任何字幕，无中文、韩文、拼音、罗马音、人物标签、标题、贴纸或水印。 Negative： 人物串脸、身份互换、双人同框、分屏、面部变形、发型服装漂移； A只有磨皮没有电影光线、全画面泛黄泛粉、过度柔焦、眼睛嘴部模糊、塑料皮肤； B仍然有浪漫散景与柔晕、B欠曝看不清脸； 所有镜头同一种表情、A也凶狠、B也娇柔、失控尖叫、持续晃头、夸张鬼脸； 嘴不动却发声、台词变旁白、韩语口型漂移、句子串组、对白重叠； 慢动作、花哨转场、闪白、黑帧、任何字幕、水印、乱码。

Create a 30-second, 1080p ultra-realistic Korean neighborhood action-comedy scene using the supplied first frame as the visual reference. Keep the same characters, faces, hair, outfits and environment consistent throughout. CHARACTERS: Young Korean woman, early 20s, long straight black hair, green cropped long-sleeve top, loose blue jeans and silver jewelry; confident, expressive and fearless. Young Korean man, early 20s, short black hair, gray button-up shirt, dark trousers and white sneakers; cocky at first, then increasingly nervous. SHOT 1 The woman walks down a quiet residential street carrying a small grocery bag. She notices the man riding a tiny electric scooter toward her way too fast. SHOT 2 He suddenly rings the bell repeatedly and gestures for her to move. She calmly steps aside, watching him pass with an annoyed expression. SHOT 3 A few seconds later, the scooter rolls back into frame by itself. The man runs after it, desperately trying to catch it. SHOT 4 The woman watches him sprint past her. She casually sticks out one foot and stops the scooter before it rolls into the street. SHOT 5 He finally reaches her, out of breath, and says, “Wait… you stopped it?” She simply looks at him and says, “You’re welcome.” SHOT 6 He grabs the scooter and tries to ride away again, but the woman notices his grocery bag has fallen open and vegetables are rolling down the sidewalk. SHOT 7 She quickly catches several vegetables before they reach the road. The man stands there embarrassed while an elderly neighbor watches and quietly laughs. SHOT 8 The woman hands everything back to him, gives him a playful “seriously?” look and walks away. SHOT 9 — ENDING The man starts riding away, then realizes his helmet is still in her hand. She raises it above her head with a tiny satisfied smile. He stops, turns around and walks back toward her as she laughs. VISUAL STYLE: Ultra-realistic Korean residential neighborhood, early-2000s MiniDV/camcorder aesthetic, handheld camera, imperfect framing, slight autofocus hunting, natural exposure shifts, subtle motion blur, soft digital detail, warm afternoon sunlight, realistic Korean houses, parked scooters, utility poles, narrow streets and authentic everyday atmosphere. Fast comedic timing, natural body language and hard cuts between shots. AUDIO: Only natural diegetic sound — footsteps, scooter motor, bell, distant traffic, birds, neighborhood ambience, vegetables hitting the pavement, natural dialogue and laughter. Absolutely no music or score. NEGATIVE PROMPT: No gore, blood, serious injuries, identity drift, outfit changes, duplicated people, extra limbs, distorted hands, teleportation, impossible physics, cartoon reactions, CGI look, subtitles, captions, text, logos or watermark.

一、任务目标 生成一段完整连续的 15 秒中国仙侠电影短片。 核心喜剧主线只有一条，并且要让观众 第一遍就立刻看懂： 一名敌方剑客故意挑拨师姐与小师妹，篡改师姐原本说过的话，让小师妹误以为师姐认为她“三招必败”。 真正反转是： 她们从来没有讨论“小师妹能不能打赢”，而是在讨论这个敌人究竟能撑几招。 二、整体风格 整体必须同时具备以下气质： 电影级写实质感 纯古风中国仙侠美学 真正史诗级高手对峙的镜头语法 一本正经的冷面喜剧 默片式反应节奏 港式动作喜剧的利落效率 清楚的“三拍递进” 明确的铺垫与回收 Arri Alexa 电影机质感 稳定清晰的面部微细节 细腻胶片颗粒 自然体积光 喜剧必须克制，不夸张，不胡闹，不靠人物变蠢。 三、人物身份锁定 角色 A｜剑仙师姐（沿用首帧中可见的身份；未出镜角色按下述描述建立） 始终保持同一个人物： 25–30 岁东亚女性 身材高挑纤细 椭圆脸 深色杏眼 黑色长发半挽 白玉簪固定 同一套白色刺绣真丝汉服 银色腰封 玉佩 白色布靴 唯一一柄银色长剑 角色 B｜小师妹（按以下描述建立并保持一致，不要求第二张图片） 始终保持同一个人物： 20–25 岁东亚女性 身形娇小 圆润灵动的脸 黑发编辫 同一套青绿色亚麻汉服 深色腰带 木簪 黑色布鞋 唯一一柄深色钢剑 其他角色 敌方剑客 一名敌方剑客 负责挑拨、误导和最终出手 不可抢走喜剧情绪中心 年长师父 同一个年长师父 只在关键时刻补最后一刀 必须保持安静、克制、一本正经 四、环境 DNA 与空间原则 以唯一首帧图中可见的环境为起点，结合下述描述建立统一的古风空间。不要求其他背景参考图。 正式构图前，先无声整合并统一以下内容： 真实地形 建筑语言 材质 植被 水体 天气 山雾 主光方向 倒影 空气纵深 在此基础上，重新规划出一个： 本轮独有、完整统一、可信的新空间。 环境规则 背景中的以下元素始终自然鲜活： 风 水体 植被 云层 远处普通弟子 空间环境声 但必须满足： 背景绝对中立 不制造包袱 不制造反转 不主动推动剧情 不替人物表达情绪 五、三镜头结构 镜头 1｜0–5s 景别 全景或远景 画面内容 同一个剑仙师姐 同一个小师妹 一名敌方剑客 一名年长师父 四人位于统一空间中，地理关系清楚稳定。 敌方剑客站在两人正前方，年长师父只在数步之外安静存在。 剧情与对白 敌人故意看着小师妹挑拨道： “你师姐亲口说，你与我交手，三招必败。” 同一个小师妹非常缓慢地转头看向师姐，问： “三招？” 同一个白衣剑仙没有任何紧张，只平静纠正： “我说的是他。” 镜头重点 第一拍要先让观众以为矛盾是“师姐看不起小师妹” 小师妹转头这一瞬必须清楚 师姐的纠正必须平静、确定、不带解释欲 这一句要直接翻第一层误会 镜头 2｜5–10s 景别 中景或牛仔景 连续性要求 保持以下内容全程稳定： 同样两名女子 同一个敌人 同一个师父 同样服装 同样长剑 完全一致的地理空间 剧情与对白 小师妹立刻重新看向敌人。 这一次她不是生气，而是真的有些惊讶，说： “三招？师姐，你也太给面子了。” 敌人刚才得意的表情当场僵住。 浅景深后方，同一个师父从头到尾没有插话，只安静地抬起一根手指。 敌人发现后，恼火问： “你什么意思？” 镜头焦点转移到师父。 师父保持绝对认真，只回答： “一招。” 完整停顿半拍。 敌人终于彻底破防，大喊： “欺人太甚！” 随后主动拔剑，向师姐正面冲去。 镜头重点 第二拍完成喜剧递进 “三招”升级成“一招” 师父这一刀必须极干 停顿要留足，让观众和敌人同时意识到自己成了被预测的对象 敌人破防后立刻进入动作段 镜头 3｜10–15s 景别 特写或极近特写 连续性要求 保持以下内容稳定： 同一个师姐 同一个小师妹 同一个敌人 同一个师父 服装、长剑、站位、空间关系全程一致 动作设计 同一个师姐始终保持放松，甚至不把自己的银色长剑完全拔出。 她只完成一个： 非常清楚 物理合理 极短促 极精准 的动作组合： 侧身闪避敌人正面冲击 用仍然带鞘的同一柄银色长剑 短促敲击敌人持剑手腕 顺势借敌人的前冲力量改变其重心 结果： 敌人的长剑旋转飞脱 安全插入旁边地面 敌人自己单膝落地 全程不得出现血腥伤害 动作后的对白 全场安静。 同一个小师妹慢慢举起一根手指，认真说道： “师父猜得真准。” 同一个剑仙师姐低头观察败下阵来的敌人，平静回答： “其实半招。” 镜头焦点再次转向师父。 师父一本正经说道： “我已经给他留面子了。” 结尾画面 极近特写中： 敌人依次看看师父、师姐、小师妹 发现自己一句都反驳不了 小师妹用力咬住嘴唇憋笑 师姐仍然非常正式地朝他行一个武者承让礼 精准切在： 敌人彻底无语的表情上。 六、表演原则 师姐 始终平静 不炫耀 不解释过度 强者松弛感要稳定 喜剧成立依靠认真，而不是玩笑腔 小师妹 反应要快 情绪不夸张 从“误会”到“听懂”再到“补刀”，节奏要清楚 憋笑感要真实克制 敌人 前半段必须真心自信 中段逐渐僵住 后段破防但不能演成小丑 笑点来自后知后觉，不来自愚蠢 师父 台词极少 每句都像盖棺定论 必须一本正经 越认真越好笑 七、动作与镜头要求 动作身体力学清晰自然 敌人冲击要真有气势 师姐应对要极省、极准、极短 最终胜负必须一眼看懂 不要复杂炫技连招 不要拖泥带水 不要光污染特效 不要随机剑气 摄影要求 16:9横屏 严格三个连续清晰镜头 运镜干净克制 前景、中景、远景保持自然视差 镜头服务人物关系与包袱回收 不做花哨炫技运动 八、声音与同步要求 原生同步普通话对白 口型精准 喜剧停顿明确 视线关系准确 拔剑、碰撞、长剑落地、归位等声音清楚可辨 环境声持续自然存在 不生成字幕 保持以下表现要求： 多模态参考连续性 多人物复杂运动稳定性 导演级运镜控制 15 秒多镜头同步音视频输出能力 九、严格避免 blurry bad quality low quality low resolution noisy jpeg artifacts watermark text error deformed mutated bad anatomy poorly drawn hands bad composition out of frame disfigured inconsistent character changing clothes face morphing background shift glitching cuts disappearing props 额外避免： 字幕 过度夸张表演 无意义搞笑表情 复杂难懂的反转 动作软绵绵 花哨慢镜 随机法术 环境制造笑点 配角抢戏 敌人突然降智 十、最终效果目标 整段短片的观感应该是： 先制造“师姐似乎看低小师妹”的表面误解 很快翻转成“原来被评价的是敌人” 最后再用动作和师父补刀，把包袱彻底坐实 最终关键词： 仙侠、冷面喜剧、高手对峙、三拍递进、误导反转、师徒补刀、半招制敌、电影级写实。

vertical 3:4 TikTok-style UGC commercial video, soft pastel aesthetic, cozy modern bedroom lighting, warm natural daylight, premium Apple-style commercial quality, smooth handheld camera movement, realistic cinematic skin texture, shallow depth of field, clean minimal Japanese room interior. Main character: cute young Japanese woman, early 20s, short dark-brown bob haircut with soft bangs, fair skin, natural makeup, expressive eyes, cozy oversized pastel blue sweater, friendly relatable vibe, looks exactly like a Japanese version of the girl in the reference storyboard, soft kawaii aesthetic, natural facial expressions. Language & Voice: She speaks English with a natural soft Japanese accent, casual Gen-Z tone, friendly and relatable delivery, natural lip sync. Video pacing: Fast TikTok cuts every 1.5–2 seconds, energetic but cozy editing, subtle zoom-ins, smooth transitions, light upbeat lo-fi background music. SCENE 1 (0:00–0:02) — PROBLEM INTRO Medium close-up shot. Girl holds tangled wired earphones with frustrated expression while sitting at desk. Dialogue: “Ahhh… cable headphones are sooo annoying…” Natural room lighting, realistic emotion, slight handheld camera shake. SCENE 2 (0:02–0:04) — PROBLEM GETS WORSE Top-view POV shot into backpack. Hands struggling with tangled cables stuck between laptop and charger. Dialogue: “Every time in my bag… it gets messy…” Quick cuts, realistic everyday frustration. SCENE 3 (0:04–0:06) — PRODUCT REVEAL Cinematic close-up. Girl opens pastel pink wireless earbuds case toward camera. Soft glowing product lighting, shallow depth of field, premium product commercial feel. Dialogue: “So I tried these cute earbuds…” Subtle slow-motion case opening. SCENE 4 (0:06–0:08) — COLOR VARIANTS Top-view aesthetic product shot. Multiple pastel earbud cases arranged neatly: pink, mint, sky blue, yellow. Hands pick up pink version. Dialogue: “The colors are super kawaii…” Soft pastel background shapes, Instagram aesthetic. SCENE 5 (0:08–0:10) — PRODUCT DETAIL Extreme macro close-up. Earbuds inside charging case, glossy reflections, smooth texture, satisfying cinematic lighting. Dialogue: “And the case is really compact.” Tiny focus pull transition. SCENE 6 (0:10–0:12) — USING EARBUDS Medium close-up. Girl smiling while putting earbuds into ears naturally. Dialogue: “Just take it out and use instantly.” Soft hair movement, cozy natural acting. SCENE 7 (0:12–0:14) — TOUCH CONTROL Close-up side profile shot. Girl taps earbud touch control while music UI glows softly on phone screen in background. Dialogue: “It even has touch control… no buttons!” Modern cinematic tech vibe. SCENE 8 (0:14–0:15) — ENDING / LIFESTYLE Medium shot. Girl studying on laptop, listening to music happily, then looks at camera and smiles while holding pink earbuds case. Dialogue: “For study, music, calls… really perfect for daily life.” Ending frame: Cute smile + product toward camera. Overlay text: “Choose your favorite color ✨” Final aesthetic: Clean pastel Japanese lifestyle commercial, cozy Gen-Z TikTok ad, realistic acting, premium wireless-earbuds commercial energy, soft cinematic color grading, smooth motion blur, ultra realistic, highly detailed.

Generate an ultra-realistic vertical 3:4 iPhone-style GRWM vlog featuring a young Japanese woman (20–24). She has fair skin, almond-shaped dark eyes, soft blush makeup, straight black hair with bangs and pastel pink streaks. Natural Tokyo influencer vibe with soft Japanese-English accent. STYLE: Fast-paced TikTok editing, handheld selfie + cinematic rear camera mix, anime-inspired realism, neon Tokyo lighting, slight motion blur, autofocus breathing, VHS overlays, Japanese text stickers, J-pop energy, authentic imperfections. INITIAL LOOK: oversized pastel hoodie, bare face skincare, hair clipped up, cozy Shinjuku apartment filled with plushies, manga, LED lights. FINAL LOOK: full Harajuku fashion (oversized jacket, layered skirt, fishnet sleeves, platform sneakers, kawaii bag, headphones), glam makeup with glitter eyes + glossy lips. CUT SCENES (15s TOTAL) 0–2s — INTRO GRWM Selfie in bedroom mirror, she pins hair. Dialogue: “Hello guys… today GRWM in Tokyo, ne~” 2–4s — SKIN PREP Close-up skincare near window, toner + moisturizer. Dialogue: “Skin prep is very important, okay?” 4–7s — MAKEUP TRANSFORM Quick cuts: glitter eyeshadow, eyeliner, lips, hair styling. Neon lighting starts. Dialogue: “I want cute Harajuku vibe today!” 7–10s — OUTFIT CHANGE Fast fashion montage: jackets, skirts, fishnets, platform shoes, mirror spin. Dialogue: “Maybe too much… but Tokyo style is always extra!” 10–12s — FINAL READY Perfume spray, headphones on, handbag grab, confident pose. Dialogue: “Okayyy, now we go Shibuya!” 12–15s — SHIBUYA ENDING Cinematic reveal at Shibuya Crossing, neon lights, crowd motion. She vlogs selfie-style. Dialogue: “See you guys in Tokyo! Bye bye~” End text: “東京 GRWM ✨”

Create a 15 second cinematic villain entrance scene in a desaturated western realism style. The main subject is a powerful blonde female villain in a dusty abandoned town, styled like a modern neo-western revenge figure. She wears a dark red tube top, fitted leather pants, silver jewelry, and boots. Her expression is cold, controlled, and revenge-driven. She walks forward with absolute confidence, as if she has returned to settle an old score. Scene structure: 0–3 sec (hook): Open on a dusty abandoned town with dry wind moving through the street. Use a low, tense trailer-style opening. Close sound detail: boots pressing into dirt, slow breathing, distant wind. Reveal her entering frame and walking forward with purpose. 3–7 sec (setup): Cut to a wider cinematic shot showing the emptiness of the town around her. Use trailer-style camera movement with a low-angle tracking shot and slight handheld grit. Dust trails lightly around her legs as she advances. Her eyes stay fixed on someone or something ahead. 7–11 sec (tension/action): Cut closer. The camera pushes in as she keeps walking. Her breathing becomes more present. She delivers a short, sharp dialogue line in a low, controlled voice, such as: “You should’ve finished this when you had the chance.” Let the line feel personal and dangerous. 11–15 sec (ending): She stops. Hold on her reaction in a close-up. Let the emotion be restrained but deadly not explosive, just final. The last 2–3 seconds should slow down and stay on her face as dust moves in the background. End with tension still hanging in the air. Camera: trailer style cinematic cuts, low-angle tracking, wide environmental framing, medium push-in, final close-up hold. Lighting: natural desaturated western daylight, dry atmosphere, dusty realism, no glossy stylization. Mood: revenge, threat, inevitability, controlled menace. Audio: heavy wind ambience, boots on dirt, steady breathing, sparse cinematic tension bed, one strong dialogue line, no abrupt ending. Ending frame: close-up hold on her reaction, emotionally loaded and dangerous.

Animate this photo realistically. The man (Charlie) speaks first saying "Pam always go Max!" with natural lip sync and mouth movements. Then the woman (Pam) replies saying "you got it Charlie!" with natural lip sync....

Ultra-realistic cinematic full-body portrait of a beautiful young woman sitting peacefully on a rocky riverside at golden-hour sunset, surrounded by lush trees and warm natural scenery. She has soft feminine facial features, luminous natural skin, expressive brown eyes, subtle natural makeup, rosy cheeks, and long dark-brown hair styled loosely in a slightly messy low bun with delicate strands framing her face. She is wearing an elegant flowing ivory-white linen blouse with rolled-up sleeves and matching wide-leg linen trousers, creating a relaxed yet sophisticated bohemian aesthetic. She sits comfortably on a textured woven blanket spread over the pebbled riverbank, with an open notebook resting across her lap and a pen naturally held in one hand. A rustic woven basket with a soft beige blanket sits beside her. Capture her in a peaceful reflective moment, sometimes gazing toward the glowing sunset over the river and sometimes turning gently toward the camera with a subtle warm smile. Golden sunlight creates beautiful rim lighting through her hair, soft highlights across her face, and shimmering reflections across the calm water. Lush foliage in the background melts into creamy cinematic bokeh. Professional luxury lifestyle photography, intimate storytelling atmosphere, warm golden-hour color grading, natural skin texture, realistic fabric details, soft shadows, atmospheric depth, shallow depth of field, creamy bokeh, 85mm portrait lens, f/1.8, HDR, photorealistic, ultra-detailed, cinematic composition, subtle film grain, natural lighting, high-end editorial photography, 8K detail, vertical 3:4 composition, same face as reference image. Animate the supplied first frame: she breathes gently, blinks, shifts her gaze toward the sunlight and smiles softly. A light breeze moves loose hair and fabric while the river ripples naturally. Slow, restrained camera push-in. Preserve her identity, pose, outfit and surroundings; no sudden pose changes, additional people or dialogue.

women flying across the city like superwoman, holding a can of Coca-Cola in hand, wide-angle lens view from the front, close-up focus on the can, city in dynamic motion blur, cinematic lighting, ultra-realistic, high speed action

Animate the supplied baseball-stands first frame as a candid live-broadcast fan-camera moment. Keep the same adult woman, face, hairstyle, clothes, blue stadium seats and nearby spectators. She looks to the side, notices the camera, blinks in surprise and lets her expression soften into a small, natural smile. Keep her seated and preserve her body proportions. Use a stable telephoto shot with a gentle zoom and realistic stadium depth of field. Nearby spectators make small natural movements. Soft crowd ambience and distant game sounds, no dialogue. No new overlay text, logos, duplicated people, face morphing or venue changes.

Create a 15-second dark gothic psychological-horror scene. Use the supplied first frame for the room, lighting and any visible character; no other reference image is required. Sin is an adult female jester with white makeup, a red nose, dark eye makeup, red twin-horn hair, a bell crown, ivory ruff and a red-and-blue court costume. Her mirror double looks identical but acts independently. Keep the antique gold mirror fixed on the right, burgundy curtain and warm sconce on the left, charcoal walls and wooden floor. Stage both faces clearly from the mirror’s left-front side. Exactly one Sin outside and one double inside at the start. 0–3 seconds: Sin approaches the mirror and adjusts her crown. The reflection follows half a beat late and keeps smiling after Sin stops. A restrained push-in reveals the mismatch. 3–5.5 seconds: Sin shakes her head, looks frightened and retreats. The double does not copy her; it leans closer, amused. Keep both opposing reactions visible. 5.5–8 seconds: A small patch of glass becomes an elastic membrane. The double’s left hand grips Sin’s right wrist and pulls. Sin braces her feet and grips the rigid frame with her free hand, visibly resisting. 8–11.5 seconds: Sin is pulled inward through the screen-left half while the double steps outward through the screen-right half. Keep their paths separate and show the double’s foot landing before it takes weight. Brief hair/fabric occlusion may hide one continuity cut. The glass seals intact; Sin is now trapped inside and cannot follow. 11.5–13 seconds: The double turns its shoulders and face toward camera. Behind it, Sin independently panics and presses her palms against the inside of the glass. 13–14 seconds: The double gives a quiet, wicked smile. Sin remains frightened and never mirrors that smile. 14–15 seconds: The double raises one finger to its lips and whispers "Shh." Sin strikes the glass in a silent plea. Cut to black. Maintain charcoal #211E1B, burgundy #4B171A, aged brass #9C7544, ivory #D8D0C1 and blue-grey #29313B. Warm left-side light, neutral makeup, restrained saturation, readable reflections. Audio: heels, bells, cloth, breath, shoe scrape, grip friction and muffled impacts. The final "Shh" is the only spoken line. No music, narration or subtitles. No shattering, merged bodies, extra limbs, duplicate occupants, face changes, position jumps, frame bending or lighting resets.