実写志向の AI 動画プロンプト

A breathtaking, high-quality, ultra-photorealistic cinematic 30-second single-shot video with smooth, continuous camera work and no cuts. The young Japanese woman from Image 1, with her signature long dark hair, wispy bangs, serene face, and warm smile, wearing an elegant white summer dress with a blue floral pattern, is the main character throughout. [0-5s] Japan Beach Golden Hour: The young woman is walking gracefully along the sandy shore of Shonan Beach during a glorious golden hour sunset. Warm, soft rim light glows on her hair and shoulders. She looks back over her right shoulder directly toward the camera, smiling warmly. The camera slowly orbits 360 degrees around her clockwise. [5-10s] Hawaii Beach Walk: As the 360-degree rotation completes, the scenery seamlessly transitions into a tropical Hawaiian beach with turquoise waters. Her facial identity, hairstyle, and smile remain perfectly consistent. She continues walking into the gentle waves of the ocean. The camera follows her, submerging smoothly under the clear water with natural motion blur. [10-15s] Majestic Undersea: Beautiful sun rays pierce through the water surface. The young woman glides gracefully and peacefully through the crystal-clear deep ocean water. Her hair and dress flow naturally with the gentle currents. Colorful tropical fish swim around her. A small vibrant fish swims upwards, and the camera follows it, rising rapidly to the surface. [15-20s] New York City: Breaking through the water surface, the scene seamlessly transforms into a bustling street in Manhattan, New York, during sunset. She continues walking elegantly along the sidewalk, her facial identity and serene expression perfectly consistent. A chic, modern light coat naturally appears over her dress. The camera tracks her smoothly as she walks toward a grand glass-walled skyscraper. [20-25s] Building Interior to Rooftop: The camera seamlessly enters the building through the glass, moving past a modern office lobby. She walks through the corridor and steps onto a spacious high-rise rooftop helipad. The camera slowly orbits her and begins to ascend vertically, revealing the beautiful city skyline scaling down below under a purple sunset sky. [25-30s] Space to Shonan (Perfect Loop): The camera passes through the soft clouds into the high atmosphere and outer space, showing a beautiful panoramic view of the Earth. The young woman floats weightlessly and peacefully in space. The camera rotates 180 degrees around her and dives gracefully back down through the clouds. A splash of water clears, returning to the original Shonan Beach. She stands in the exact same position, looking back and smiling warmly at the camera, forming a perfect, seamless loop.

微距电影镜头，夜晚的厨房台面被拍成像一片巨大的金属荒原。面包屑、盐粒和香草碎片像岩石和灌木 scattered 在台面上。画面主角是一只奇特的“胡椒研磨器螳螂”：身体是一只小型银色不锈钢胡椒研磨器，竖直偏前倾，底部和两侧长出细长真实的螳螂腿，前方有两条标志性的镰刀状捕捉前肢。它没有正常昆虫头部，主体就是胡椒研磨器本身，行动谨慎而机械感十足。 一只油亮的棕色蟑螂从左侧的餐巾纸阴影后缓缓爬出，快速接近胡椒研磨器螳螂。空气紧张，像一场深夜厨房决斗。胡椒研磨器螳螂突然停住，身体微微抬起，像在锁定目标。接着它猛地转动顶部研磨头，发出清脆机械声，大量黑胡椒碎末像烟雾弹一样从顶部和前方喷洒出来，直接打向蟑螂。 蟑螂被胡椒粉呛到，慌乱后退，触角疯狂摆动，脚步打滑，迅速倒退逃回左侧阴影。空气中飘起细小黑胡椒颗粒 and 轻微粉雾。胡椒研磨器螳螂停在原地，前肢高高举起，像一个冷静获胜的守夜战士。 时间线分镜 0–3秒： 超近距离微距镜头。银色胡椒研磨器螳螂在厨房台面上缓慢行走，周围有面包屑、盐粒和香草碎片。金属表面有柔和反光，螳螂腿动作真实自然。 3–6秒： 左侧阴影里，一只棕色蟑螂突然出现，快速逼近胡椒研磨器螳螂。镜头轻微跟随，营造紧张感。双方在一小片空地上短暂对峙。 6–7秒： 胡椒研磨器顶部突然快速转动，发出机械摩擦声。大量黑胡椒粉末猛地喷出，形成浓密颗粒状黑色喷雾，正面打向蟑螂。 7–10秒： 蟑螂被胡椒粉刺激后慌乱倒退，迅速逃回左侧阴影并消失。空气中漂浮黑胡椒粉尘。胡椒研磨器螳螂保持静止，前肢抬起，像胜利者一样守在原地。

One singular drip of water falls from the faucet landing in the cup of water creating heart shaped ripples heart shaped ripples

【生成模式】 Seedance 2.0 Fast 【视频规格】 严格生成 15 秒视频，16:9 横屏，三个清楚镜头。 【整体风格】 诗意仙侠视错觉喜剧，真实电影高清质感，采用古典中国留白构图、克制的冷面表演、清晨柔光、低饱和玉绿色与象牙白色调、细腻胶片颗粒，并借鉴巴斯特・基顿式空间错觉，通过可控的摄影机旋转揭示真相，而不是依赖法术失败制造笑点；全片为纯古代中国世界，不出现任何现代元素。 【角色锁定】 使用 @图片 1 和 @图片 2 严格锁定角色身份、面部、发型、身材比例、服装、鞋子与饰品。 角色 ID A｜剑仙师姐 25–30 岁的东亚女性，椭圆脸，白皙自然肤色，深色杏眼，黑色长发半挽，以白玉簪固定，身材高挑纤细，穿白色刺绣真丝汉服、半透明分层宽袖、银色腰封、玉佩 and 白色布靴。 角色 ID B｜小师妹 20–25 岁的东亚女性，圆润灵动的脸，黑发编辫，身形娇小，穿青绿色亚麻汉服、深色腰带、木簪 and 黑色布鞋。 【镜头 1｜0-5s｜全景或远景】 画面整体有意顺时针旋转 90 度，让一面陡峭山壁在视觉上变成水平地面。 角色 ID A：剑仙师姐，25–30 岁的东亚女性，椭圆脸，白皙自然肤色，深色杏眼，黑色长发半挽，以白玉簪固定，身材高挑纤细，穿白色刺绣真丝汉服、半透明分层宽袖、银色腰封、玉佩 and 白色布靴。她从鞋到头完整出现在画面中，看似双手负后、从容不迫地沿垂直绝壁笔直向上行走，头发和衣袖自然垂向画面中的万丈深渊。 角色 ID B：小师妹，20–25 岁的东亚女性，圆润灵动的脸，黑发编辫，身形娇小，穿青绿色亚麻汉服、深色腰带、木簪 and 黑色布鞋，看似正抓着下方一棵古松。 【镜头 2｜5-10s｜中景或牛仔景】 同一个穿青绿色亚麻汉服的小师妹仰头问道：“师姐，这就是踏壁登天？” 同一个穿白色刺绣真丝汉服的剑仙师姐继续平静前行，淡然回答：“心正，天地自平。” 背景始终保持同一面山壁、同一棵古松、同样的漂浮山雾、石台 and 远处瓦顶水榭；必须维持错误的垂直方向，但不能改变人物身体结构。 【镜头 3｜10-15s｜特写或极近特写】 摄影机平稳逆时针旋转 90 度，恢复真实水平线，清楚揭示同一个穿白色刺绣真丝汉服的剑仙师姐其实只是在一面矮墙旁的平坦石台上普通行走，而同一个穿青绿色亚麻汉服的小师妹只是随意靠在古松旁。 小师妹面无表情地说道：“师姐，天地平了，镜头歪了。” 极近特写：剑仙师姐努力维持的端庄微笑瞬间凝固，一侧眉毛轻轻抽动，随后若无其事地回答：“这是考你眼力。” 最后以一声干涩古琴拨弦 and 一颗松果落在石板上的声音收尾。 【技术要求】 严格总时长 15 秒，16:9 横屏，三个清楚镜头，原生普通话对白，人物面部与服装稳定，场景几何关系连续，山雾、真丝衣料、头发、古松 and 重力方向真实一致，不生成画面字幕，始终只出现两名可见角色。 【负面词】 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

Vertical 9:16 handheld phone video, 15 seconds, one unbroken take, no cuts.Close, high, looking down onto a miniature snow village. White powder snow with tiny wooden chalets and a frozen lake on the right.

Astronaut on an ice moon under a giant ringed world. And at the end… a shooting star.

Style: Photorealistic cinematic realism. A cramped, atmospheric late-night Tokyo ramen shop — steam curling off broth pots, red paper lanterns, worn wooden counter, dangling bulb lighting, condensation on the windows, rain-slicked street glowing outside, subtle handheld camera movement, rich analog film grain. 0–4s: Medium-wide shot. A young woman in an oversized denim jacket sits alone at the counter, composed with the faintest hint of amusement, slowly slurping noodles with chopsticks. Behind her, a young cook crosses the narrow kitchen carrying two steaming bowls of ramen in one hand and a full kettle of hot broth in the other. Near the entrance, an elderly regular begins to rise from his stool to leave. 4–8s: Dynamic tracking shot. The elderly man stumbles backward and collides heavily with the cook. The bowls, noodles, broth, and kettle explode upward into the air. Broth stretches into long steaming ribbons with suspended droplets, noodles frozen mid-arc, steam curling in slow motion. The camera immediately performs a smooth orbital move around the collision as time freezes at the exact peak of the spill. Every face is locked in an expression of shock. Only the young woman at the counter continues moving naturally, calmly lifting another bite to her mouth. 8–17s: Slow 360-degree orbit. The camera gracefully circles the frozen shop. Broth hangs in midair like molten glass ribbons and floating spheres, noodles suspended in delicate loops, steam frozen into soft drifting clouds. The bowls, kettle, and chopsticks drift weightlessly in suspended time. The woman takes another slow bite, her eyes half-closed with an almost bored expression. Then the frozen world begins to rewind elegantly. Every droplet, every noodle, and every person moves backward with perfect precision, returning toward their original positions. 17–24s: Medium shot. The rewind ends flawlessly. The cook is once again carrying the bowls and kettle steadily. The woman looks up and casually raises two fingers, softly calling the cook by name. The cook turns his head at exactly the right moment, stepping around the elderly man as he rises. The collision never happens. A subtle, private smile flickers across the cook's face. 24–30s: Extreme close-up. Hard cut to the woman's face. She takes one final slow slurp with a knowing smile, her eyes almost completely closed. Very shallow depth of field. The red lanterns behind her melt into soft, warm bokeh. Visual Quality: Photorealistic, cinematic image quality with highly accurate fluid and steam simulation, realistic gravity and collisions, motion blur only on moving objects, stable character consistency, seamless cinematic temporal continuity, natural facial performances, and premium filmic color grading.

In Namibia’s hush, where the desert drinks rain, millions of lilies rise from the dust. Pink and white, they drink the last light, their faces turned to the sky they once knew only as a star. For five brief days they bloom, a quiet riot against the dry. And when the sun finally touches them, they answer with fire.

First-person handheld cinematic walk through a rain-washed Asian night market, capturing neon reflections, steam, and cozy crowd atmosphere. First-person walker (unseen POV) (Camera holder at eye level, moving forward through the market with natural footsteps and slight handheld sway; hands/body not visible), Night market vendors (Several stall cooks wearing aprons, working behind steaming food stations, flipping skewers and stir-frying in woks), Crowd of shoppers (Mixed local pedestrians and visitors in the foreground and background, some carrying transparent umbrellas, moving casually) and Neon signage (Red, amber, and pink glowing signs hanging from stalls and overhead wires, with Chinese/Japanese-style characters, reflecting off wet pavement). The first-person camera walks steadily forward through the bustling night market; vendors cook and steam rises from food stalls while pedestrians drift past on both sides, and umbrellas catch neon light as rain droplets glisten mid-air in Narrow covered night market lane after a light rain, wet reflective asphalt, dense food stalls on both sides, warm neon glow overhead, steam banks drifting into the walkway. The visuals feature Cinematic, cozy, atmospheric; warm neon reds/ambers/pinks reflecting on wet surfaces, soft haze from cooking steam, shallow depth of field with glistening highlights. Use First-person handheld POV at eye level, forward tracking with natural slight sway and footsteps, occasional subtle focus pulls between foreground pedestrians and background neon signs; continuous single shot. Audio includes Ambient market sound: sizzling oil, clinking spatulas, distant chatter, soft rain patter on umbrellas, low warm background music blending into the crowd noise. Maintain consistency: Maintain strict first-person POV throughout — no visible head/hands; Keep the single continuous walking shot without cuts; Preserve wet pavement and neon reflection visibility in every frame; Retain steam from food stalls as an atmospheric layer; Do not introduce drone shots, tripod static shots, or POV breaks.

WORLD: Midnight shipping shipyard in the pouring rain, palette of deep shadow black, concrete gray, and crisp white, one vivid glowing-crimson accent, volumetric floodlights, floating rain droplets, layered depth of field. CHARACTERS: - OPPONENT: A massive

7-shot handheld phone vlog with snapshot realism, slight tilt, natural handheld shake, warm indoor lighting, fine film grain, photorealistic. A woman Image1 eats lunch alone at a rustic Korean samgyupsal restaurant. Image1 defines only her face and hairstyle. She wears a loose olive-green cotton short-sleeve shirt tucked into dark trousers. She remains seated on the same side facing the camera throughout. Only her face is visible; the grill attendant appears only as hands. A wooden table holds a charcoal grill, tongs, scissors, ssamjang, garlic, sesame oil, lettuce, and napkins. Thick pork belly sizzles as smoke rises under warm pendant lights. Natural, subtle reactions throughout. Shot 1 (0–2s): Hands place pork belly on the grill. She watches the sizzle and says, "오 소리 봐." Camera gently pulls back. Shot 2 (2–4s): Overhead shot as she cuts the pork with scissors on the grill. Slight handheld shake. Shot 3 (4–6s): She makes a lettuce wrap with pork, garlic, and ssamjang, takes a bite, closes her eyes, and says, "음~~ 대박." Shot 4 (6–8s): She grabs a hot piece with chopsticks, blows on it, eats, and smiles: "So good it's dangerous." Shot 5 (8–10s): She wipes her fingers and mouth while watching the grill continue to sizzle. Shot 6 (10–13s): A hand places a frosty beer (or soju). She lifts the glass toward the camera and says, "캬 이거지." Shot 7 (13–15s): She takes a long sip, sets the glass down, leans back with a satisfied smile as the grill continues sizzling. Audio: No music. Only authentic restaurant ambience—grill sizzling, scissors, tongs, chewing, glass sounds, soft chatter, and her natural dialogue. No subtitles, text, logos, watermarks, or duplication of the reference image.

Handheld home-video style montage, 7 continuous shots with varied angles (avoid a single camera angle or one-take). Filmed one-handed on a phone with slight handheld shake, natural outdoor daylight, subtle film grain, photorealistic realism. A woman Image1 repots a houseplant alone on a cozy apartment balcony. Image1 is used only for her face and hairstyle. She wears a faded denim short-sleeve shirt over a white tank top with a dirt-smudged canvas gardening apron. The narrow balcony has potted plants, terracotta pots, potting soil, gardening tools, string lights, and warm midday sunlight. She is the only person in the video. The plant goes from being root-bound in a cracked plastic pot to freshly repotted and watered. Dialogue is natural spoken Korean. Shot 1 (0–2s): She taps and squeezes the old plastic pot to loosen the plant. "자, 나와라~" Shot 2 (2–4s): Close overhead as she removes the plant, exposing tightly coiled roots and gently loosening them. Shot 3 (4–6s): She places it into a terracotta pot, fills it with fresh soil by hand, and says, "훨씬 낫다." Shot 4 (6–8s): She brushes dirt from the leaves, straightens the plant, and says, "There we go." Shot 5 (8–10s): She wipes her dirty hands on her apron and smiles with a small approving nod. Shot 6 (10–13s): She waters the plant with a small metal watering can. Soil darkens as it absorbs the water. "쭉쭉 마셔." Shot 7 (13–15s): She steps back, leans on the balcony railing, admires the plant, wipes her brow, and exhales contentedly. Audio: No music—only natural sounds of tapping plastic, crumbling soil, rustling roots, patting soil, fabric movement, pouring water, light breeze, birds, and a quiet satisfied breath. No subtitles, text, logos, or watermarks. Do not recreate or copy the reference image itself.

TITLE BURJ KHALIFA VLOG FORMAT 15 seconds | 16:9 | Live-action travel vlog Fictional dramatization. Documentary-style visual realism. SETTING The Burj Khalifa observation deck in Dubai. Floor-to-ceiling glass windows overlooking the entire Dubai skyline with visible Sheikh Zayed Road, Downtown Dubai, the Dubai Fountain and surrounding skyscrapers. Time: golden hour approaching sunset. Warm sunlight, clear atmosphere, gentle indoor ambience and distant city activity. CHARACTERS A small group of friends visiting the Burj Khalifa observation deck. Maintain consistent clothing, accessories, bags and character identities throughout the sequence. Natural tourist behaviour throughout the experience. STORY 00–04s — ESTABLISHING Wide documentary shot reveals the Burj Khalifa observation deck as visitors admire the panoramic skyline. The camera slowly walks toward the glass while Dubai stretches endlessly into the distance. Golden sunlight reflects naturally across the city. People move casually without acknowledging the camera. 04–08s — EXPLORATION Cut to a handheld vlog perspective following the group as they walk along the observation deck. One person points toward famous landmarks while another captures photos with a smartphone. Natural conversations, laughter and genuine reactions. The camera operator follows with realistic footsteps, subtle framing corrections and natural motion blur. 08–12s — HIGHLIGHT The group reaches the panoramic viewing area. One friend records a selfie video with the skyline filling the background while others admire the sunset. The camera circles naturally to reveal the immense height and scale of the city below. Wind lightly moves clothing near the outdoor terrace. No exaggerated cinematic movements. 12–15s — ENDING The camera settles beside the viewing glass. The group quietly watches the sunset as Dubai's city lights begin illuminating below. One person smiles toward the skyline before walking away. End on a calm, memorable travel moment. CAMERA Grounded travel-documentary cinematography. Mix: wide environmental coverage medium handheld following shots brief close reaction shots Natural handheld instability. Realistic operator movement. Plausible tourist camera positions. No impossible drone transitions. No excessive slow motion. LIGHTING Lighting originates entirely from the natural environment. Warm golden-hour sunlight entering through the observation windows. Natural exposure adapts between bright exterior views and interior spaces. Realistic highlight roll-off. Detailed shadows. No artificial rim lighting. PHYSICS Real gravity and momentum. Natural walking movement. Believable clothing motion. Hair responds subtly to breeze on the outdoor terrace. Realistic reflections on glass. Authentic interaction with railings, viewing telescopes and observation deck surfaces. HUMAN PERFORMANCE Relaxed, believable tourist behaviour. Natural smile

Skateboarding city streets POV, 15s. The video should end with a wall featuring a doodle that says “yes&

A 15-second photorealistic smartphone selfie vlog that looks exactly like a real mobile phone recording. A young woman (image = her face and hair) is exploring the quiet streets of Kyoto, Japan on a bicycle during a sunny afternoon. She films herself in selfie mode with natural handheld movement, subtle hand shake, realistic smartphone autofocus, authentic stabilization, and true-to-life colors. Smiling at the camera, she says, "Let's explore Kyoto today!" She happily rides through narrow traditional streets lined with wooden houses, then notices she's taken a wrong turn and laughs, saying, "I think I'm lost!" Continuing down a peaceful path, she unexpectedly discovers a beautiful hidden Japanese shrine surrounded by tall trees and red torii gates. She parks her bicycle, walks toward the shrine with a surprised smile, captures a few quick shots of the peaceful scenery, then turns the camera back to herself and says, "Getting lost was worth it!" She gives a warm wave, smiles brightly, and says, "See you next time. Bye!" before reaching toward the phone to end the recording. The video should feel exactly like a genuine smartphone vlog with natural human movement, consistent facial features, realistic bicycle riding, authentic Kyoto atmosphere, no beauty filters, no CGI, no AI-plastic appearance, no subtitles, no logos, no watermark, and no background music—only birds, bicycle sounds, soft wind, footsteps, and natural outdoor ambience.

Use the uploaded reference image as the exact character reference. Preserve her facial identity, hairstyle, eye color, makeup, skin tone, body proportions, and facial consistency throughout every shot. Maintain the same character consistently. Create a 15-second ultra-realistic GRWM office routine with 15 fast-paced cinematic shots (≈1 second each), smooth match cuts, whip pans, object transitions, premium lifestyle cinematography, and realistic motion. 1. Alarm rings as she opens her eyes. 2. She sits up, stretches, and smiles. 3. Quick skincare splash in the bathroom. 4. Applies light makeup in front of the mirror. 5. Brushes and styles her hair. 6. Changes into a chic office outfit. 7. Puts on heels, watch, and earrings. 8. Grabs her handbag, laptop, and coffee tumbler. 9. Locks the apartment door behind her. 10. Walks confidently through the apartment lobby. 11. Hails a cab on a busy city street. 12. Rides through downtown with cinematic city reflections. 13. Steps out in front of a modern glass office building. 14 Scans her access card and enters the office. 15 Sits at her desk, opens her laptop, smiles at the camera, and begins working. Ultra-realistic commercial quality, smooth handheld and gimbal camera movement, natural expressions, realistic body movement, premium office interiors, luxury apartment, cinematic color grading, shallow depth of field, realistic reflections, 4K HDR, 16 9, no text, subtitles, logos, or watermarks.

【总体要求】 12秒写实系单次换装短视频，9:16竖屏。严格复刻参考视频《平衡换装》的动作逻辑：女主单腿平衡，用抬起脚的外侧面托住“鞋履＋一堆衣服”，将整堆衣物挑向空中，迅速收腿蓄力，只出一次高速踢腿并准确命中；脚与衣物重合的同一帧瞬间换装。 变装必须是一帧硬切，不是渐变、变形或衣服逐渐覆盖身体。无台词、无旁白、无字幕、无水印，保留背景音乐。 【人物锁定】 女主为25岁以上成年女性，使用@1 。面部五官、肤色、发型、身高与身形比例全程一致，不换脸、不变年龄、不改变体型。动作具有真实重心、肌肉发力、呼吸起伏和轻微平衡晃动。 【服装锁定】 变装前：黑色短袖T恤、淡紫色微喇长裤、黑白运动鞋。 变装后：女主瞬间换上预设#1 的完整造型，严格保持#1的颜色、版型、材质和搭配。全片只变装一次，换装后的造型持续到结尾，不增加、不删除、不替换任何服装、鞋履或配饰。 【起始衣物包】 开场道具不是单只鞋，而是预设#1造型对应的鞋履与一堆折叠衣服叠在一起，形成紧凑、具有真实重量的衣物包。鞋履轮廓位于衣物堆上方，折叠衣料从下方托住鞋，部分布料边缘自然下垂。 整堆鞋履和衣服被女主左脚运动鞋的外侧面、外侧脚背和脚踝附近从下方托住。道具不放在脚尖，不穿在脚上，不悬浮。挑起后，鞋与衣服必须作为同一个组合一起飞行，不能分散成多个物体。 【场景】 简约现代住宅室内，白墙、浅灰色硬质地面，背景保留白色衣柜、房门、洗衣机和少量生活用品。自然窗光混合柔和室内光，家具、背景和光线方向全程不变。真实手机视频质感，皮肤保留自然纹理，衣物和鞋履材质清楚。 【近距离固定机位】 固定紧凑全身景，摄影机距离女主约2.1—2.3米，高度约1.05米，约30—32mm等效焦段，接近平视并带轻微仰角。 女主站立时约占画面高度80%，位置略偏右，为左侧抬腿和衣物包抛物线预留空间。头顶保留约8%空间，脚底保留约4%地面。人物、支撑脚、抬起的左腿和空中衣物包始终完整入镜。全程固定，不推拉、不摇移、不跟拍、不切换特写。 【严格时间分镜】 [00:00-00:00.90] 女主侧身站立，右腿作为唯一支撑腿，右膝微屈；左腿向侧面抬起至接近水平。左脚略微向内转，使运动鞋外侧面朝上，像一个倾斜的小托盘。 预设#1对应的鞋履和折叠衣服堆叠成紧凑组合，被左脚外侧面稳稳托住。女主侧头注视脚侧的整堆衣物，注意力完全在衣物包的平衡状态上。双臂自然抬起，右脚脚踝、腰腹和肩膀进行真实的小幅调整。她不看镜头。 [00:00.90-00:01.05] 女主只用左脚脚踝和脚外侧做一次短促、清晰的向上挑动，将鞋履与整堆衣服同时挑向斜上方。 这只是一次轻巧的挑物动作，不算正式踢腿。衣物包整体离开脚侧，沿一条短促抛物线上升。衣料边缘受到空气影响轻微展开，但鞋履和主体衣服仍聚拢在一起，不散开。伴随轻微“嗖”声。 [00:01.05-00:01.45] 衣物包快速上升并接近最高点。女主立即收回左腿，左膝迅速弯曲靠近胸腹，脚跟收向身体，形成紧凑蓄力姿态。 她的身体向右侧轻微倾斜，右腿独立支撑，双臂随重心变化自然展开。头部与眼睛始终追踪整堆衣物，不看镜头，没有停顿摆拍。 [00:01.45-00:01.65] 衣物包越过最高点并开始下降，所有鞋履和衣服仍沿同一条抛物线运动。女主完成收腿蓄力，停顿不超过0.2秒，准确判断衣物包的中心落点。 [00:01.65-00:01.85] 女主只进行一次正式高踢腿，并且第一次便准确成功。 左腿从屈膝状态爆发式向斜上方伸直，膝关节迅速打开，小腿和脚尖在约0.2秒内完成完全伸展。出腿飞快、干净、利落，带有自然运动模糊和衣料破风感，不使用慢动作。 左脚准确迎向正在下落的整堆鞋履和衣服，不踢空、不减速、不修正、不补踢。 [00:01.85，瞬间换装硬切] 左脚与衣物包中心第一次重合的这一帧，立即进行无缝动作匹配硬切。 接触前一帧：女主仍完整穿着黑色T恤、淡紫色长裤和黑白运动鞋；预设#1对应的鞋履与衣物包仍在空中。 接触后一帧：女主已经完整穿上预设#1造型，严格保持#1的颜色、版型、材质和搭配；空中的鞋履、衣服和配饰同时彻底消失。 中间不存在任何过渡帧。禁止出现半边旧装、半边新装；禁止衣服从皮肤表面生长；禁止布料缠绕身体；禁止颜色扩散或透明叠化。变装就是前后两个完整状态之间的一帧硬切。 切换前后女主的左腿高度、伸展角度、身体倾斜、手臂位置、面部、背景和光线完全一致。音乐重拍同步响起一声低沉有力的“砰”。 [00:01.85-00:02.80] 换装完成后，女主保持高踢腿顶点约0.1秒，再控制左腿沿原轨迹下降。她先看向已经换装完成的左脚，目光随腿部下降。 左脚平稳落地后，她先确认站姿，再缓慢抬头看向镜头。神情从专注自然转为从容自信，嘴角轻微上扬。 [00:02.80-00:04.10] 女主摆出第一个海妖风时尚姿态：身体形成柔和S形曲线，一只手扶腰，另一只手缓慢掠过头发和肩侧。她先看向镜头侧方，再缓慢转回目光，下巴微抬，露出神秘、自信的浅笑。 预设#1造型保持稳定，严格保持#1的颜色、版型、材质和搭配。 [00:04.10-00:05.40] 她转换成三分之二侧身姿态，前后脚自然交错，肩部向后舒展，腰部轻微转动。她直视镜头挑眉、轻轻歪头，随后缓慢眨眼，表现含蓄、调皮的挑逗感，保持高级时装展示尺度。 [00:05.40-00:06.60] 她顺势缓慢转身，将背部朝向镜头。头发沿肩背自然滑动，预设#1造型不发生变化。转身完成后，她越过肩膀回头看向镜头，一侧嘴角微微扬起。 [00:06.60-00:09.30] 女主背对镜头向房间深处优雅行走3—4步，采用从容的时装走走秀步态。双脚自然交替，肩膀放松，手臂随步伐轻微摆动，身体产生克制、流畅的节奏。 摄影机保持固定，以完整背影全身景记录她的步态和整套#1造型，不跟拍、不放大、不使用身体局部特写。 [00:09.30-00:10.40] 她在背景前停下，一只脚向侧后方轻点地面，仍保持背向镜头，随后越过肩膀回头。她轻轻挑眉，露出神秘侧笑，并做一次短促的俏皮眨眼。 [00:10.40-00:12.00] 结尾高潮：音乐最后一个强拍到来时，她突然做一次干净利落的半转身，同时甩动头发，向镜头方向迈出一步，停在三分之二正面姿态。 她一只手扶腰，另一只手抬到下巴附近，下巴微扬，以直接、自信、具有海妖神秘感的目光凝视镜头。头发和服装边缘在强拍上形成自然运动弧线，最后0.3秒定格收束。 【音频】 00:00—01:85为空灵女声哼唱、高音与轻微混响，保留支撑脚摩擦声、衣料轻响、挑起衣物包的“嗖”声和高速出腿破风声。 01:85脚与衣物包重合时，音乐瞬间进入强劲电子鼓点Drop，并同步一声低沉“砰”重击音。 01:85—12:00持续电子节拍，保留鞋底落地声、行走脚步声、衣料摩擦声和甩发破风声。结尾加入上扬收束音。无对白、无旁白、无字幕。 【强制负面约束】 禁止渐变换装、溶解、叠化、衣服逐渐生成、布料爬上身体、颜色扩散、旧装慢慢消失；禁止任何中间变装状态。必须在脚与衣物包第一次重合时完成一帧硬切：前一帧百分之百旧装，后一帧百分之百预设#1完整造型。 禁止擅自修改#1；禁止改变#1的颜色

A cinematic first-person travel vlog following a cheerful young woman wearing a maroon jacket as she explores Japan during golden hour. The video opens with a handheld selfie shot in front of a famous Japanese landmark as she smiles naturally at the camera. Smooth cinematic transition to a cozy café where she enjoys a latte by the window. Cut to a peaceful walk through a lush city park with cherry trees gently swaying in the breeze. Transition to a vibrant street food market where she samples local snacks while colorful lanterns glow overhead. End with a modern Tokyo metro ride as she gazes out the window, city lights reflecting on the glass. Natural candid expressions, realistic human motion, stabilized handheld camera, soft golden-hour lighting, warm cinematic color grading, shallow depth of field, subtle lens flares, ultra-realistic documentary travel aesthetic, 4K, highly detailed.

A 30-second photorealistic smartphone travel vlog. A young Japanese woman wearing a pastel yukata explores a traditional Japanese summer festival at night. She films herself in selfie mode with natural handheld movement, realistic phone camera quality, and authentic ambient sounds. She walks through lantern-lit streets, buys fresh takoyaki and smiles after taking a bite, plays a traditional festival game and wins a small plush toy, then watches a spectacular fireworks display with the crowd, capturing her genuine excitement. She ends the vlog by waving at the camera and saying, "Thanks for coming with me. See you next time. Bye!" The video should feel natural, cinematic, and realistic with consistent appearance, smooth human motion, and authentic festival atmosphere.

7秒，9:16，一镜到底，甜美古风OOTD短视频。7秒音频完整播放，配合人物舞蹈。 严格以@人物参考图人物设定图中的成年女性为唯一人物：以右侧脸部特写作为人物长相参考，以左侧正面图 and 中间背面图作为服装、发型及背部结构参考。完整保持她的小巧鹅蛋脸、清秀五官、红色眉心花钿、深棕黑色高盘古典发髻、浅金色花叶发簪、珍珠流苏和两侧垂落珠链；完整保持浅香槟白与淡桃粉古风舞衣、刺绣抹胸内搭、浅杏色透明长袖纱衣、浅金色腰封、金属链饰、多层垂坠长裙。全程禁止换脸、换装、改变发型或增减饰品。 仅参考@视频人物的动作轨迹、姿态、节奏、景别变化和构图，不继承原视频人物的脸、发型、服装、颜色或材质。纯白无影棚，高机位轻微俯拍，超广角轻微透视，镜头基本固定，人物始终保持全身或大半身清晰入镜。 0–1秒，人物从画面右侧提着少量裙摆轻快走入，长裙和薄纱袖自然摆动； 1–2秒，人物靠近镜头，上身轻微前倾，抬眼注视镜头，神情清甜克制； 2–3秒，双手轻扶腰间，形成自然收腰姿态，俏皮微微歪头； 3–4秒，优雅侧身转过半圈，展示服装背面、盘发和腰封结构，再回眸看向镜头； 4–5秒，转回正面浅浅微笑，双臂向身体两侧自然舒展，轻纱袖缓慢垂落； 5–6秒，双手轻放身后，先低头整理姿态，再微微歪头看向镜头； 6–7秒，再次轻扶腰间，人物先扭头看向与镜头呈90°角的侧前方，视线先聚焦在镜头外某个远方，然后再转向镜头，以甜美、自信、端庄的古风姿态定格结束。 动作自然连贯、轻盈克制，不做大幅度舞蹈动作。裙摆、薄纱衣袖、珠链流苏和耳侧发丝随动作产生轻微真实惯性，避免夸张飞舞。明亮柔光，清透高亮，柔和浅香槟色调，古装影视广告与韩系时尚短片结合的高级质感，节奏轻快，无对白，搭配轻盈空灵的国风流行音乐。人物身份、五官、发型、服装纹样和饰品位置全程稳定，面部清晰稳定，真实皮肤纹理，4K质感，细节清晰。 不要使用左侧正面图中的空白遮脸作为人物面部，人物长相必须来自右侧脸部特写；不要继承视频中原人物的脸、发型或服装；不要新增人物，不要换装，不要跳切，不要镜头大幅移动或抖动，不要快速旋转，不要剧烈甩袖，不要裙摆遮脸，不要肢体变形、重复手臂、多余手指、手部粘连、穿模、服装突变、饰品消失、发型变化或背景变化。

A cinematic 15-second vertical morning routine video of a young woman in a modern luxury home. The video begins with her waking up in a cozy bedroom illuminated by warm golden sunrise light. She reaches to the bedside table, picks up her smartphone, checks the time, smiles softly, and stretches naturally. She walks into a bright minimalist bathroom and brushes her teeth while looking into the mirror. Next, she performs her skincare routine in front of the mirror, gently cleansing her face and applying toner, serum, and moisturizer with graceful, realistic movements. Finally, she walks into a stylish dining room, sits at a wooden breakfast table with toast, fresh fruit, coffee, and orange juice, takes a sip of coffee, and smiles while looking out the sunlit window. Warm golden-hour lighting, soft shadows, natural expressions, realistic body movement, elegant camera motion, shallow depth of field, photorealistic skin, luxury lifestyle aesthetic, cinematic color grading, ultra-realistic, 4K HDR, smooth transitions, premium commercial quality, vertical 9:16.

15s handheld home-video vlog, 7-shot montage. Photorealistic phone footage with slight tilt, natural shake, window light, subtle film grain. A woman (use Image1 only for facial identity and hairstyle) does laundry alone on a quiet morning. Outfit: oversized cream linen shirt with rolled sleeves, grey knit shorts, loose cotton apron. Cozy sunlit laundry nook with an open front-load washer, overflowing basket, wooden drying rack, clothespins, and warm sunlight. She is the only person in the video. Sequence: untangles wet clothes → shakes out a shirt → checks a collar stain by the window → hangs it saying "Good enough." → finds a mismatched sock → struggles with a heavy bedsheet while laughing → finishes hanging it and quietly admires the sunlit laundry. Dialogue is natural spoken Korean (except "Good enough"), reacting casually to each moment. Ambient sound only: washer winding down, wet fabric, clothespins, rustling clothes, soft laughter, breeze. No subtitles, text, logos, or watermarks. Do not recreate or copy the reference image—use it only for facial identity and hairstyle.

Create a 15-second premium UGC-style product advertisement for the Hamilton Beach 2-Way Coffee Maker shown in the reference image. Use the reference image strictly for product identity, design, proportions, colors, buttons, carafe, red travel mug, and overall appearance. Do not redesign or alter the product. The video should feel like a real creator filmed it at home with a high-end smartphone, natural lighting, realistic handheld camera movement, authentic human interaction, and premium but believable visuals. No overly polished commercial look, no artificial CGI feeling. Scene 1 — 0:00–0:02 Close-up of a hand pressing the Program button on the coffee maker. The digital display is clearly visible. Morning sunlight enters through a kitchen window. Subtle handheld camera movement. Voiceover: “Okay, this has seriously upgraded my mornings.” Scene 2 — 0:02–0:04 Medium-wide shot of the coffee maker sitting naturally on a modern kitchen counter. The red travel mug is positioned on the single-serve side and the glass carafe is on the other side. A young woman casually prepares her morning coffee in the background. Voiceover: “I love that I can brew it exactly how I want.” Scene 3 — 0:04–0:06 Macro close-up of fresh ground coffee being placed into the filter basket. Show realistic coffee grounds, texture and small natural movements. Voiceover: “Just add my favorite coffee…” Scene 4 — 0:06–0:08 Extreme close-up of freshly brewed coffee flowing smoothly into the glass carafe. Capture realistic steam, liquid movement and rich coffee color. Warm morning lighting. Voiceover: “…and let it do its thing.” Scene 5 — 0:08–0:10 The woman pours the freshly brewed coffee into a clear glass cup, then takes a natural first sip. Genuine relaxed expression, no exaggerated acting. Voiceover: “And the coffee comes out so good.” Scene 6 — 0:10–0:12 Quick lifestyle shot showing the single-serve side and full carafe side together. Camera slowly moves around the machine, highlighting its stainless-steel finish and compact design. Voiceover: “Plus, I can make one cup or a full pot.” Scene 7 — 0:12–0:14 Beautiful hero shot of the coffee maker on the kitchen counter with the red travel mug beside it, soft sunlight, subtle steam rising from the coffee. Camera slowly pushes in toward the product. Voiceover: “Honestly, it just makes mornings easier.” Scene 8 — 0:14–0:15 Final close-up of the woman holding her coffee while the coffee maker remains visible in the background. Natural smile. End on a clean product-focused frame. On-screen text: “Better Coffee. Better Mornings.”

Use the storyboard grid Image1 as the exact visual reference for character design, apartment layout, clothing, handbag, props, lighting, and composition. Animate the six storyboard frames as one continuous 15-second cinematic sequence. Maintain the character’s exact appearance throughout. Maintain the apartment as one fixed physical environment. Do not alter furniture, architecture, kitchen layout, décor, handbag design, clothing, hairstyle, or character proportions. Movement should feel like polished feature-film 3D character animation: natural body weight, realistic walking rhythm, subtle secondary movement in hair and clothing, expressive facial animation, believable hand interaction with objects, and smooth cinematic camera movement. Jarvis is an AI voice only. He is never physically shown. Character sheet is Image2 Apartment exterior and bag is Image3 ⸻ SHOT 1 — 0:00–0:02.5 — 2.5 SECONDS Begin from inside the apartment facing the entrance. The front door opens and the character walks naturally into the apartment after returning from work. She carries the black Hermès Birkin 25 by its two short top handles in one hand and her car keys in the other. Her walk communicates end-of-workday fatigue: slightly lowered shoulders, relaxed pace, calm face. She looks tired but comfortable now that she is home. As she crosses the doorway, she reaches back naturally and begins pulling the door closed behind her. The handbag responds naturally to her walking motion with subtle weight and momentum. No long strap appears. Camera: Wide eye-level establishing shot. Very subtle cinematic push-in as she enters. Keep the established entry architecture visible. Audio: Door opening, soft heel footsteps, subtle handbag leather movement, quiet apartment ambience. Dialogue: None. ⸻ SHOT 2 — 0:02.5–0:04.5 — 2 SECONDS Cut to the entry console. the character walks up to the console without fully stopping. She extends the hand holding her car keys and casually releases them onto the wooden surface. The keys fall naturally and land with a distinct metallic clink. Her other hand continues holding the black Birkin by its short handles. She barely looks at the keys. The action is automatic and habitual. Immediately after releasing them, her attention shifts toward the living room and kitchen. Camera: Medium side-angle shot. Keep Jema, the Birkin and the console visible. A subtle focus shift can follow the keys as they land before returning to the character. Audio: Footsteps followed by a crisp key CLINK. Dialogue: None. ⸻ SHOT 3 — 0:04.5–0:07 — 2.5 SECONDS the character continues from the entry area into the living room, passing behind the cream sofa on her way toward the kitchen. Without stopping, she swings the Birkin slightly forward and casually tosses it onto the cream sofa. The handbag leaves her hand with believable weight, travels a short distance and lands on the cushion. The cushion compresses slightly beneath it. The bag settles naturally and remains there. As she releases the handbag and continues toward the kitchen, the character speaks casually: the character: “Jarvis, update me on my work today.” Her delivery is relaxed and familiar. She does not call loudly and does not look for a speaker or device. She talks to Jarvis as naturally as she would speak to someone already in the room. Her lips must synchronize accurately with the dialogue. Camera: Wide tracking shot. Follow Jema smoothly from the living room toward the kitchen while keeping the established apartment geography readable. Audio: Soft heel footsteps, subtle handbag landing on sofa. the character voice: Young adult woman, confident, slightly tired after work, natural conversational delivery. ⸻ SHOT 4 — 0:07–0:09.5 — 2.5 SECONDS the character reaches the kitchen and moves behind the island onto the kitchen side of the counter. She reaches toward the completely empty clear drinking glass on the countertop. The glass must remain completely empty throughout this shot. She has not poured water yet. Her eyes briefly settle on the glass as she picks it up or positions it naturally for pouring. She listens to Jarvis without interrupting her routine. Jarvis begins responding through the apartment: JARVIS (O.S.): “Miss Jema, today Claude connected to PayBox Agent by MoonPay…” Jarvis should sound like a sophisticated male AI assistant: calm, polished, articulate, composed and subtly warm. Not robotic and not overly emotional. the character does not speak. Her expression remains neutral and attentive. Camera: Medium shot from the living-room side of the island looking toward Jema behind the counter. Gentle cinematic push-in. Audio: Jarvis voice integrated naturally into the room ambience. Very subtle glass contact with countertop. ⸻ SHOT 5 — 0:09.5–0:12.5 — 3 SECONDS Cut to a cinematic close-up of the character’s hands at the same kitchen counter. There is a clear water bottle and empty glass on the counter, She picks up the clear water bottle and begins pouring water into the same previously empty glass. Show a natural continuous stream entering the glass and the water level gradually rising. Jarvis continues seamlessly: JARVIS (O.S.): “…and traded. Your portfolio is up by 17 percent…” Jema continues pouring normally. Then Jarvis finishes: JARVIS (O.S.): “… that was under ten minutes.” As he says “under ten minutes,” Jema’s pouring motion hesitates very slightly. Do not make her freeze dramatically. It should be a tiny physical reaction showing that the number caught her attention. Camera: Tight close-up of bottle, hands, water stream and glass. Very subtle camera drift. Shallow depth of field. Audio: Clear natural water-pouring sound underneath Jarvis’s voice. ⸻ SHOT 6 — 0:12.5–0:15 — 2.5 SECONDS Cut to the character‘s face and upper body behind the kitchen island. She stops pouring and holds the now-filled glass naturally. For a brief beat, she processes what Jarvis just said. Her eyes glance slightly upward. Her eyebrows rise subtly. Then the smallest impressed smile forms at the corner of her mouth. Keep the reaction understated. No huge eyes, dramatic gasp, head jerk, or exaggerated cartoon reaction. She is impressed, but AI automation like this is already part of her life. End on this subtle expression, setting up Grid 2. Camera: Medium close-up with a slow, almost imperceptible push toward her face. Shallow depth of field with the established kitchen softly blurred behind her. Audio: Water pouring stops. Brief room ambience after Jarvis finishes. No dialogue from Jema. ⸻ ANIMATION CONTINUITY The six shots are one continuous sequence. the character enters with keys + Birkin → keys stay on console → Birkin lands and stays on sofa → the character reaches kitchen → glass is empty before pouring → water is poured → the character ends holding the glass. Keep her movement restrained and natural. No unnecessary gestures, no duplicates, random head turns, excessive blinking, floating objects, hand deformation, prop morphing, wardrobe changes, hairstyle changes, or apartment changes. Total duration: exactly 15 seconds. 16:9 landscape. Storyboard Grid 2 - Video Prompt (15 seconds) Use the storyboard grid @Image1 as the exact visual reference for character design, apartment layout, clothing, handbag, props, lighting, and composition. Animate the six storyboard frames as one continuous 15-second cinematic sequence. Maintain the character’s exact appearance throughout. Maintain the apartment as one fixed physical environment. Do not alter furniture, architecture, kitchen layout, décor, handbag design, clothing, hairstyle, or character proportions. Movement should feel like polished feature-film 3D character animation: natural body weight, realistic walking rhythm, subtle secondary movement in hair and

Create an ultra-realistic premium luxury eyewear commercial featuring the same young East Asian woman throughout the entire sequence. Maintain identical facial features, shoulder-length soft ash-brown wavy hair, glowing skin, body proportions, and outfit in every shot. She wears a fitted black crop top, black athletic shorts, white sandals, delicate silver jewelry, and stylish sunglasses resting on her head. Bright tropical beach, golden-hour sunlight, crystal-clear ocean, palm trees, premium luxury fashion campaign aesthetic, cinematic camera work, soft warm tones, realistic skin rendering. The video opens with her smiling warmly at the camera while standing barefoot near the shoreline, holding a pair of premium designer sunglasses beside her face. The lenses catch the golden sunlight as she softly says, "See every moment in style." A cinematic macro hero shot showcases the sunglasses resting on warm beach sand surrounded by seashells and sparkling water droplets. Close-up shots highlight the premium frame, polished hinges, elegant lenses, and luxurious craftsmanship. She gracefully puts on the sunglasses while maintaining eye contact with the camera. She strolls along the beach as gentle waves wash over her feet. Smooth tracking shots follow her while the ocean breeze naturally moves her hair. She occasionally adjusts the sunglasses, smiles confidently, and looks back toward the camera. She relaxes beneath palm trees, sipping fresh coconut water and laughing with friends. Cinematic close-ups capture the sunglasses reflecting the ocean, palm trees, and glowing sunset while emphasizing her effortless summer style. As golden hour approaches, she walks into the shallow waves, playfully splashing water while the sunset reflects across the ocean. Slow-motion shots capture sparkling water droplets, flowing hair, and beautiful reflections across the lenses. She sits quietly on the beach watching the sunset before taking one final selfie with the ocean behind her. The sunglasses remain the hero accessory, complementing the warm golden atmosphere. Final hero shot: she walks confidently toward the camera, removes the sunglasses, then gracefully places them back on while smiling. She says, "Style begins with the perfect view." The camera slowly pulls back to reveal the glowing beach, ocean waves, palm trees, and golden sunset as the sunglasses become the focal point. Natural ambient audio only: ocean waves, gentle breeze, birds, footsteps on sand, distant beach ambience, water splashes, leaves rustling, and soft laughter. No background music, no subtitles, no logos, no watermarks, and no on-screen text. Premium luxury fashion cinematography, ultra-realistic skin texture, physically accurate lighting, shallow depth of field, realistic reflections, smooth handheld and gimbal movement, 16:9 widescreen, 4K HDR.

Style: Ultra-photorealistic, luxury commercial, cinematic, one continuous tracking shot, premium fragrance boutique, warm ambient lighting, realistic human movement, 35mm lens, shallow depth of field, HDR, 8K, smooth gimbal stabilization. Scene 1 (0–3s) – Entering the Store A stylish young man wearing a fitted black t-shirt and dark jeans walks into a luxurious perfume boutique. The camera follows from a front-left tracking angle as he confidently enters. Elegant wooden shelves filled with premium glass fragrance bottles glow under warm LED lighting. Polished white marble flooring reflects the lights, creating a premium atmosphere. Scene 2 (3–6s) – Walking Through the Aisle The camera smoothly tracks beside him as he walks slowly through the fragrance aisle, casually looking at the collection. Hundreds of luxury perfume bottles are neatly arranged on illuminated shelves. Soft reflections on the glass bottles, clean modern interior, cinematic depth of field, realistic body movement, natural walking pace. Scene 3 (6–10s) – Browsing the Collection The camera moves slightly closer while he turns his head toward the shelves, appreciating the premium fragrances. Close-up shots capture elegant perfume bottles with golden liquid, crystal-clear glass, metallic caps, soft bokeh in the background, warm luxury retail lighting, realistic facial expressions. Scene 4 (10–13s) – Hero Ending The camera slowly pushes in as he continues walking past the display with confidence. The luxury boutique stretches into the background, showcasing rows of illuminated perfume shelves and a premium showroom. Smooth cinematic motion, elegant atmosphere, luxury fashion commercial aesthetic, ultra-realistic details, clean composition, high-end color grading, no text, no watermark, no logos.

A photorealistic smartphone selfie vlog that looks exactly like a real mobile phone recording. A young woman (image = her face and hair) visits a cozy thrift store on a sunny afternoon, filming everything herself with a handheld smartphone in selfie mode. The video features natural hand shake, realistic autofocus, slight exposure changes, authentic smartphone stabilization, and true-to-life colors. Smiling at the camera, she says, "Today's thrift store challenge—I'm only buying the first cute thing I find!" She walks through colorful racks of vintage clothes, shelves of accessories, plush toys, and home décor until she spots a random adorable oversized hat. Laughing, she immediately tries it on in front of a mirror, making funny poses and bursting into genuine laughter at how silly she looks. She decides to buy it, walks out of the store holding a small shopping bag, proudly shows her surprise find to the camera, smiles brightly, waves, and says, "See you in the next vlog. Bye!" before reaching toward the phone to stop the recording. The video should feel completely real with natural human movement, consistent facial features, realistic hand interactions, authentic thrift store lighting, no beauty filters, no CGI, no AI-plastic appearance, no subtitles, no logos, no watermark, and no background music—only real store ambience, footsteps, quiet conversations, clothing rustling, and natural environmental sounds.

10秒竖屏9:16，美式街头写实反转广告短剧。iPhone原生相机第一人称跟拍，摄影者像路人一样从后方追赶@1女士。自动曝光、自动白平衡、自动对焦，未经调色，无滤镜，无美颜。真实步行起伏、追赶抖动、呼吸感、转身运动模糊 and 轻微对焦延迟。 美国城市普通商业建筑外侧，灰色金属墙、红砖墙脚、狭长人行道、碎石绿化带和草坪。背景保持干净，不出现任何品牌招牌。 【人物锁定】 唯一现场真人为预设人物@1女士 穿着 图片_20260719093342_1234_183 。脸、发型、身材和整套预设服装严格参考@1，全程保持一致。 @1女士前半段始终沿着人行道向前走。第一次被拍肩和说出警告时都不能完全停下，只能稍微放慢速度、扭转上半身回头。她只能在第二次被拍肩后正式停住。 摄影者始终不露脸，只允许一只成年路人的右手短暂入镜。手掌只能轻拍@1女士的上背部或肩膀。 【手持广告牌】 广告牌不是墙面招牌，而是一块可以用双手举起的横向手持硬板。 尺寸约60厘米宽、42厘米高、厚度约5毫米，黑色轻质硬质泡沫板，明显小于@1女士的上半身宽度，可以完整藏在她身后。 正面严格使用用户提供的黑金广告参考图 hf_20260723_023253_17d01f18-b189-45de-be9c-f30db6d27116 “JOHN87445528” 广告图中的人物只是平面印刷肖像，不能变成现场真人。 她的左手自然藏在身后，暗中握着广告牌，但广告牌被身体完全遮住，广告牌不能提前露出任何边缘、金色图案或文字。 【严格时间分镜】 00:00—00:01.30 @1女士背对镜头，沿着建筑外墙旁的人行道持续向前走。她步伐自然，头发和衣服随步伐轻微摆动。 第一人称摄影者在后面加快脚步追赶。镜头出现明显但真实的上下起伏，能听见逐渐加快的脚步和呼吸声。 摄影者追到她身后，右手从画面右下方伸出，轻轻拍一下她右侧肩膀。 00:01.30—00:03.50 @1女士没有停下来。她只是稍微放慢脚步，身体继续向前移动，同时从右肩方向扭转上半身，回头盯住镜头。 她眉头收紧，用右手食指向后指着追赶的摄影者，边走边用普通话严肃警告： “你再碰我一下，就知道我是谁了！” 她不能正面站定，双脚必须继续交替向前走。警告结束后，她收回手指，把头转回前方，恢复原来的步行速度。 00:03.50—00:05.10 @1女士继续背对镜头向前走，没有等待摄影者，也没有回头偷看。 摄影者停顿半秒后再次加快脚步追上去。镜头产生更明显的追赶抖动，右手再次伸向她的同一个肩膀位置，轻轻拍下第二次。 00:05.10—00:06.40 第二次被拍后，@1女士先向前多走一步，然后突然把一只脚稳稳踩住，第一次彻底停下。 她迅速转身面对镜头。摄影者受到惊吓，本能地连续后退两步，画面产生明显晃动并扩大到全身中景。 @1女士转身的同时，左手从背后把横向广告牌突然抽出。 广告牌必须沿着她左侧腰部真实滑出，不是凭空生成。她右手立即接住广告牌右下角，双手用力一翻，将广告牌正面朝向镜头举到胸前。 抽牌、翻面、双手接住三个动作连贯快速。硬板带有真实重量和惯性，停止时轻微晃动一次。 00:06.40—00:08.30 @1女士双手稳稳举着广告牌。刚才严肃的表情突然变成自信、亲切、略带得意的商业笑容。 她从广告牌上方看着镜头，用普通话高调宣布： “我是JOHN八七四四五五二八的官方广告大使！” 说话时广告牌始终正对镜头，不左右旋转，不遮挡品牌文字。 00:08.30—00:10.00 @1女士保持原地站立，双手将广告牌稍微向镜头前递，形成自然的广告展示动作。 镜头从她的脸缓慢下移到广告牌，自动对焦锁定黑金图案和“JOHN87445528”。最后至少1.5秒保持广告牌完整清晰，她从广告牌上方露出满意笑容，画面结束。 【声音】 只有真实同期声：室外微风、远处车辆、连续步行声、摄影者追赶的加速脚步声、轻微呼吸声、两次拍肩声、衣物摩擦声、广告硬板抽出和翻转的声音，以及@1女士的中文原声。 无背景音乐，无旁白，无字幕，无笑声音效。 【硬性限制】 第一次被拍肩后，@1女士不能停下、不能正面站着说台词，必须边走边回头警告；警告结束后必须扭头继续向前走。 只有第二次被拍肩后，她才能完全停住、转身并抽出广告牌。 广告牌不能提前出现，不能凭空生成，不能变成墙面招牌、电子屏、纸张或超大广告板。 彻底禁止小狗、猫及其他动物。全片只有@1女士一名现场真人。 不要字幕、标题、水印、媒体Logo、平台界面或额外文字；广告牌上的“JOHN87445528”除外。

Cinematic vertical 9:16 beauty vlog style, 4k resolution, ultra-realistic textures, clean lighting, and smooth camera stabilization. ​Scene & Timeline Choreography ​00:00 - 00:02 ​Visual: A young woman with short, platinum-blonde hair and a glowing complexion looks directly into the camera with an excited expression inside a bright, modern white-tiled bathroom while holding a small yellow rubber duck under her chin. ​Voice Over: (Playful, surprised intake of breath or light chuckle) ​00:02 - 00:05 ​Visual: She smiles warmly while looking down at the yellow rubber duck in her hand, then looks back up toward the camera with a playful expression. ​Voice Over: "Okay, wait..." ​00:05 - 00:07 ​Visual: She brings a bottle of orange skincare serum up close to the camera, showcasing the label clearly while speaking with an enthusiastic expression. ​Voice Over: "I literally cannot believe how good this is." ​00:07 - 00:08 ​Visual: A close-up shot of her hands dispensing a drop of the translucent, glossy serum onto her fingertip and rubbing it gently against her palm. ​Voice Over: (Soft ambient bathroom sounds / gentle product texture application) ​00:08 - 00:10 ​Visual: Close-up of her face as she gently pats and applies the serum onto her cheek, highlighting her radiant skin and smooth application. ​Voice Over: "Look how it just melts in..." ​00:10 - 00:13 ​Visual: Medium shot showing her full outfit—an off-the-shoulder white long-sleeve top, dark brown leather shorts, and white thigh-high stockings—standing by the bathroom counter with plants and a towel, gently touching her glowing cheek. ​Voice Over: "...my skin already feels insane." ​00:13 - 00:15 ​Visual: She holds up the orange serum bottle again, flashing a wide, radiant smile directly at the camera while speaking excitedly. ​Voice Over: "Okay, I'm obsessed. I'm not going back!"

Create a cinematic 3D architectural visualization of a traditional Korean Hanok house on a technical blueprint background. Show a curved dark Giwa tiled roof, exposed wooden beams, raised stone foundation, Hanji sliding doors, and a wooden balcony. Add blue architectural schematics, measurements, floor plans, and cross-section overlays. Use a smooth camera rotation from side view to front view, with a colorful rainbow scanning light effect highlighting the roof and wooden joints. Warm interior lighting creates a peaceful, elegant atmosphere while showcasing the structure and design details.

CAMERA / LOOK: Handheld mini DV camcorder footage filmed by the subject herself. Slight hand shake, occasional focus hunting, imperfect framing, natural zoom adjustments, soft tape-like image quality, subtle grain, realistic auto-exposure shifts from bright kitchen morning light. Natural skin tones, mild motion blur, authentic consumer camcorder aesthetic. STYLE: Cozy coffee-prep vlog with gentle ASMR elements. Relaxed pacing, minimal dialogue, candid moments. Focus on satisfying sounds: bean grinder whirring, portafilter tamping, steam wand hissing, cup clinking, milk frothing. SUBJECT: Young man in his mid-20s, plain t-shirt, hair slightly tousled, minimal accessories. Calm, focused energy while making his morning coffee. SETTING: Small kitchen counter with an espresso machine on a bright morning. Natural daylight, coffee beans and a mug nearby, quiet atmosphere. STORYBOARD: → (3s, propped medium shot) Places camera on the counter, switches on the machine. "Morning coffee, the proper way." → (3s, overhead shot) Grinds fresh coffee beans, fine grounds falling into the portafilter. → (3s, close-up) Tamps the grounds down firmly and evenly. → (3s, handheld shot) Locks the portafilter into the machine. "Here we go." → (3s, detail shot) Espresso streams slowly into a small cup. No dialogue. → (3s, medium shot) Pours cold milk into a small steel pitcher. "Time for the milk." → (3s, macro shot) Steam wand hissing as it froths the milk. → (3s, propped shot) Pours frothed milk carefully over the espresso, forming light layers. → (3s, warm ending shot) Holds the finished cup, takes a small sip, satisfied smile. "That's exactly what I needed." → (3s, final shot) Reaches toward camera, still holding the cup. "See you later." Hand covers lens as recording ends. AUDIO NOTES: Natural kitchen ambience — grinder whirring, tamping, steam hissing, milk pouring should be clearly audible. Dialogue quiet and casual. REALISM NOTES: Authentic body language, natural blinking, genuine focused smiles, occasional careful pauses while pouring, imperfect framing, focus breathing, bright morning light shifts. Should resemble a genuine personal coffee vlog on a consumer camcorder, not a commercial or AI-generated production.

Create a 15-second seamless architectural space renovation animation. The camera remains completely fixed in the exact same position, angle, focal length, perspective, and composition throughout the entire video. No camera push-in, pull-out, panning, rotation, zoom, handheld shake, or reframing. Room dimensions, walls, windows, doors, ceiling height, and structural layout remain completely unchanged throughout the transformation. 0–3 seconds: The scene begins as a rough, unfinished interior with exposed cement walls, raw concrete flooring, exposed ceiling, visible construction materials, and an empty space. Dust settles naturally while the environment remains static. The floor installation begins, gradually spreading from left to right until the entire concrete floor transforms into premium microcement or elegant large-format light grey tiles. 3–6 seconds: Walls and ceiling begin transforming simultaneously. Rough cement walls smoothly become perfectly finished creamy-white walls. Wood veneer panels, decorative wall mouldings, built-in storage cabinets, TV feature walls, suspended ceiling structures, recessed lighting channels, and hidden LED strip lighting appear naturally in their correct architectural positions as though installed by invisible craftsmen. 6–9 seconds: Electrical fixtures, doors, window frames, skirting boards, switches, sockets, and architectural details complete themselves with realistic construction logic. Ambient lighting gradually powers on, filling the room with soft, warm illumination while maintaining a natural daylight balance. 9–12 seconds: Furniture and soft furnishings assemble sequentially. Luxury sofas, coffee tables, dining furniture, rugs, curtains, indoor plants, decorative artwork, lighting fixtures, books, cushions, and decorative accessories naturally move into place piece by piece, resembling a real construction and interior installation time-lapse. 12–15 seconds: The renovation reaches its final premium interior design. Every object settles perfectly into position while the lighting subtly adjusts to create a luxurious atmosphere. The completed room remains perfectly still for the final second, allowing viewers to appreciate the finished transformation. Overall style: Ultra-realistic architectural visualization, photorealistic materials, premium luxury interior design portfolio quality, realistic construction time-lapse, physically accurate lighting, smooth seamless transitions, natural installation logic, cinematic realism, highly detailed textures, clean modern luxury aesthetic, 8K quality.

FPV shot: The camera races through a dense forest at sunrise, weaving between towering trees covered in morning mist. Golden rays pierce the canopy as billions of floating pollen particles and tiny droplets of dew drift through the air. The camera enters a shaft of sunlight where the illuminated particles synchronize naturally, forming "MORNING" before exploding into a cascade of sparkling golden dust. Epic sunrise, cinematic nature, ultra realistic atmosphere, breathtaking VFX.

【生成模式】 Seedance 2.0 Fast 【视频规格】 严格生成 15 秒视频，16:9 横屏，三个干净清晰的镜头。 【整体风格】 电影级写实质感，采用经典冒险电影式视觉误导、默片式冷面节奏、暴风雨海岸氛围、克制人物表演、清楚的前后景空间调度，以及由摄影机运动推动的尺度揭示；喜剧来源于两位角色以过度严肃的英雄姿态处理一个无害物体，并在结尾追加第二次视觉反转。 【角色】 角色 ID A｜剑仙女性 使用 @图片 1 作为角色 ID A 的严格身份参考：同一个 25–30 岁东亚剑仙女性，椭圆脸、自然白皙皮肤、锐利深棕色眼睛、黑色长直发、高挑纤细身材，穿白色布靴、飘动的白色刺绣真丝汉服、半透明层叠宽袖、银色腰饰，佩戴玉簪并携带银色长剑。 角色 ID B｜单车女性 使用 @图片 2 作为角色 ID B 的严格身份参考：同一个 25–30 岁东亚单车女性，圆润而富有表情的脸、棕色短马尾、黄色防风外套、蓝色牛仔裤、白色运动鞋、黑色头盔和同一辆自行车。 【核心道具】 同一柄银色长剑、同一辆自行车、一条红色长绳、一只巨型充气龙风筝，以及一只体积更大的粉色小猪风筝。 【镜头 1｜0-5s｜低角度全景】 傍晚狂风中的海边楼顶，湿润混凝土地面、金属护栏、被风扯动的警示带、远处高楼、黑色云层、海雾和猛烈阵风共同建立环境。 一条巨大的黑色飞龙轮廓在城市天际线上方疯狂挣扎，雷电照亮它展开的双翼，并传来低沉咆哮。 角色 A 从鞋到头完整入镜，拔出银色长剑；角色 B 在她身旁稳住自行车，两人神情凝重地仰头观察。 摄影机缓慢稳定推近；配雷声、衣料拍打声、车链轻响和低沉管弦鼓点。 【镜头 2｜5-10s｜中景侧面跟拍】 同一个穿白色刺绣真丝汉服的剑仙女性将同一柄银色长剑射向空中，同一个穿黄色防风外套的单车女性快速踩动同一辆自行车，绷紧一条连接 “飞龙” 的红色长绳。 同一个楼顶、统一风向、金属护栏、红绳 and 城市天际线始终保持一致。 银色长剑准确勾住怪物翅膀上的绳圈；两人同时发力向后拉，白色布靴和自行车轮胎在湿润地面上真实滑动，巨龙逐渐下降。 【镜头 3｜10-15s｜揭示特写进入极近特写】 摄影机跟随坠落的怪物移动，随后快速升高并向后拉开，揭露所谓 “巨龙” 其实是一只缠在楼顶天线上的巨型充气龙风筝。 同样两位女性小心将风筝放到地面，同时松了一口气。 画外传来小孩声音：“阿姨，我的是粉色小猪！” 两人缓慢转头。 焦点转移到她们身后，一只体积更大的粉色小猪风筝正随风越飘越远。 极近特写两人同步变得呆滞的表情；银色长剑默默自动回到剑鞘。 【技术要求】 严格总时长 15 秒，16:9 横屏，三个镜头，物体运动轨迹连贯，人物面部与服装稳定，强风、绳索张力、湿润地面、风筝面料、头发和真丝衣料物理效果真实，原生同步生成普通话对白与空间声音，不生成画面字幕。 【负面词】 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

Create a 15-second ultra-realistic cinematic fantasy film featuring an original ancient red dragon protecting the last human child in a destroyed kingdom. Focus on emotional storytelling, realistic character movement, dramatic lighting, detailed environments, realistic physics, and seamless cinematic transitions. 0:00–0:03 | The Last Child A 7-year-old human child walks alone through the ruins of an ancient kingdom at night, holding a tiny glowing lantern. Smoke and ash drift through the destroyed streets while distant flames illuminate the ruins. 0:03–0:06 | The Threat Several enormous dark creatures emerge from the smoke and surround the frightened child. The child slowly steps backward as the creatures approach. 0:06–0:10 | The Dragon Appears A gigantic ancient red dragon suddenly lands between the child and the creatures. Its massive wings spread wide, golden eyes glowing intensely, as it releases a powerful roar that echoes through the ruined kingdom. 0:10–0:13 | The Protection The creatures retreat in fear. The dragon gently lowers one enormous wing around the child, shielding them from danger. The child looks up with tears in their eyes, realizing the dragon is protecting them. 0:13–0:15 | Emotional Ending The dragon looks at the child before turning toward the burning horizon as dawn begins to rise. The child gently places a small hand against the dragon’s face. Warm sunrise light breaks through the smoke as the camera slowly pushes in. Visual Style: Epic Hollywood fantasy, photorealistic dragon, realistic child, cinematic volumetric lighting, highly detailed scales, atmospheric smoke, dramatic camera movement, shallow depth of field, realistic physics, 4K, emotional storytelling, seamless transitions. Camera: Wide establishing shot → low-angle creature reveal → dramatic dragon landing → close-up of dragon and child → emotional final shot. Audio: Distant wind and fire, creature growls, powerful dragon roar, cinematic orchestral buildup, emotional soft ending.

Photorealistic cinematic 1950s American diner, chrome stools, red vinyl, neon glow and checkerboard floor, shot with modern lived-in realism and soft natural window light. Subtle handheld texture, warm practicals, rich period detail, heavy film grain. 0-4s: [Medium Wide] A striking young woman in her early 20s sits alone at the counter, calm and slightly amused, slowly sipping a tall thick milkshake through a straw. Behind her a young waitress in classic uniform approaches with a tray of eggs and bacon in one hand and a full glass coffee pot in the other. An older lady starts rising from a nearby booth. 4-8s: [Dynamic Tracking] The older lady collides hard into the waitress. Tray, plate, eggs, bacon and coffee pot explode upward in chaotic slow motion. Coffee erupts into long liquid ribbons and perfect suspended droplets. Camera immediately begins a smooth continuous orbit around the impact. Time locks completely at the peak of the spill. Every face freezes in pure shock. Only the girl at the counter keeps moving, completely unfazed. 8-17s: [Slow 360° Orbital] Camera glides in a full elegant orbit through the frozen diner. Coffee hangs in mid-air as glassy ribbons and spheres with perfect volume and surface tension. Bacon strips, eggs and the spinning tray float weightlessly. Patrons and waitress remain locked in startled expressions. The girl at the counter takes one slow, deliberate sip, eyes half-lidded, almost bored, while the entire frozen world (except her) begins an elegant reverse: every droplet, every piece of food and every person rewinds smoothly back to the exact starting positions. 17-24s: [Medium Shot] Rewind lands perfectly. Waitress stands balanced again with tray and coffee pot. The girl lifts her eyes, raises two fingers in a small casual gesture and softly calls the waitress by name. The waitress turns toward her just before the older lady begins to stand, completely avoiding the collision. A tiny private smile crosses the girl’s face. 24-30s: [Extreme Close-Up] Hard cut to her face as she takes one last slow sip. Soft knowing smile, eyes almost closed in quiet satisfaction, like she has done this a hundred times. Shallow depth of field, creamy bokeh of the neon diner behind her. Photorealistic, ultra-detailed fluid physics, perfect motion blur only on moving elements, stable characters, cinematic lighting, heavy natural film grain, no artifacts, movie-level temporal coherence, high rewatch value.

Photorealistic cinematic daytime amusement park roller coaster, bright hard sunlight, strong wind, realistic motion blur on tracks and background, subtle handheld camera energy, rich skin detail, heavy natural film grain. 0-5s: [Medium Shot] A young woman in her early 20s sits in the front seat of a moving roller coaster car, calm and slightly bored, hair whipping in the wind. Beside her sits a middle-aged man wearing what looks like a normal full head of hair. Behind them, another car with a woman is visible. The coaster drops and banks hard. 5-9s: [Dynamic Tracking] The force of the turn rips the man’s wig free. It peels off his bald head in a chaotic upward arc, spinning and flying backward through the air. The wig lands messily on the head of the woman in the car behind. Everyone’s faces freeze in pure shock and confusion at the peak of the chaos. Time locks completely. Only the young woman in the front keeps moving. 9-15s: [Slow Orbital + Detail] Camera slowly orbits the frozen cars. The wig hangs mid-air in a twisted shape with individual hairs suspended. The bald man’s scalp is fully exposed, mouth open. The woman behind is frozen mid-scream with the wig draped over her face. The young woman looks sideways, rolls her eyes and mouths “joder, otra vez”. She reaches into her pocket, pulls out a stick of chewing gum, unwraps it and calmly puts it in her mouth, starting to chew while the entire frozen scene (except her) begins a precise reverse. 15-20s: [Medium Close-up] The rewind is controlled and elegant: the wig lifts off the woman behind, flies backward through the air in reverse, and returns exactly to the moment it is only beginning to peel off the bald man’s head. Time freezes again at that precise instant — the front edge of the wig just lifting, a few strands already loose. 20-26s: [Tight Medium] Still frozen for everyone else, the young woman takes the chewed gum out of her mouth, reaches over and firmly presses it onto the center of the bald man’s scalp, right under the lifting wig. With the same hand she smooths and presses the wig back down into perfect place, locking it with the gum. She sits back, looks straight ahead with a tiny private smile, completely unbothered. 26-30s: [Medium Shot] Time suddenly resumes at full real-time speed. The coaster continues its drop. The man touches his head, feels the wig still firmly in place, looks confused for a second, then breaks into a relieved, happy smile. The young woman stares forward, already chewing a new piece of gum, expression of quiet satisfaction. Photorealistic, ultra-detailed wind and hair physics, perfect motion blur only on moving elements, stable characters, cinematic lighting, heavy natural film grain, no artifacts, movie-level temporal coherence, high rewatch value.

FORMAT 15 seconds | 16:9 | Live-action mountaineering survival drama Fictional dramatization. Documentary-style visual realism. SETTING K2 during a summit push above Camp II. Steep snow-covered ridge with exposed rock faces, fixed ropes, deep ice walls and extreme alpine terrain. Altitude above 6,500 meters. Time: early morning. Thin air, sub-zero temperatures, powerful winds, drifting snow and distant mountain echoes. CLIMBERS A small expedition team of experienced high-altitude climbers ascending together on fixed ropes. Maintain consistent climbing suits, helmets, harnesses, oxygen equipment, backpacks and character identities throughout the sequence. Movement reflects realistic fatigue and the physical effects of extreme altitude. STORY 00–04s — ESTABLISHING Wide documentary shot reveals the climbing team carefully ascending a narrow exposed ridge above Camp II. The immense slopes of K2 dominate the background while fixed ropes stretch upward into the clouds. Snow blows continuously across the ridge. 04–08s — TENSION Cut to a handheld perspective following directly behind the climbers. Ice axes strike firmly into hard snow while crampons bite into the icy surface. Heavy breathing is audible through oxygen masks. One climber briefly looks uphill as subtle cracks appear in the snow above. 08–12s — CLASH A massive slab avalanche suddenly fractures high above the ridge. The snowfield collapses naturally before accelerating downhill. Climbers instinctively attempt to anchor themselves to the fixed rope as a powerful wall of snow engulfs the ridge. The camera is violently swept into the avalanche, tumbling with realistic momentum, rapidly alternating between snow, sky, ice and rock while visibility disappears beneath dense powder. Avoid graphic injuries or gore. 12–15s — AFTERMATH The camera lies partially buried beneath fresh snow, tilted toward the shattered ridge where pieces of rope and scattered climbing equipment remain visible. Fine snow falls across the lens while distant wind echoes through the mountains. End on unresolved tension. CAMERA Grounded expedition-documentary cinematography. Natural handheld instability. Realistic operator movement. Plausible helmet-mounted and shoulder-level perspectives. No impossible drone transitions. No excessive slow motion. LIGHTING Lighting originates entirely from the natural alpine environment. Cold early-morning sunlight diffuses through thin clouds and blowing snow. PHYSICS Real gravity and momentum. Climbers remain physically attached to fixed ropes until realistic force overwhelms anchors. Snow fractures, accelerates and behaves according to believable avalanche dynamics. Loose snow responds naturally to wind. HUMAN PERFORMANCE Restrained, believable behaviour under extreme stress. Heavy breathing from altitude. Careful foot placement. Constant balance corrections. Natural communication through brief gestures and shor

POV — Morning Commute in Coastal Japan | Ultra-Realistic, One Continuous Take First-person POV, no cuts, one unbroken handheld-feeling shot from our own eyes on the way through a small seaside town. Ultra-photorealistic, phone-in-hand realism — natural camera micro-shake, subtle head bob with each step, occasional glance down at our own hands and shoes. 4K, shallow but natural depth of field, real morning light, no color-grading fantasy, no neon. Beat 1 — Descending into the station (0–4s): We walk down worn concrete station stairs into a small coastal-town metro entrance, one hand loosely gripping a canvas bag strap that swings into the lower frame. Fluorescent tint mixes with daylight bleeding from the entrance behind us. Commuters pass in soft focus; we hear muffled announcements and the shuffle of footsteps. Beat 2 — The orange cat (4–9s): We slow and crouch — the horizon dips as our POV lowers toward the ground. An orange tabby sits by a ticket-gate pillar, tail curled. Our hand (natural skin texture, slightly bitten nails, cuff of a sleeve) reaches into frame and gently strokes its head. The cat leans into the touch, eyes half-closing, ears flicking. We hold there a beat, warm and quiet. Beat 3 — Boarding, settling by the window (9–14s): We rise, the horizon lifts, and we step through the train doors just as the chime sounds. POV turns and moves toward a window seat; we sit, and the frame settles against the glass. A faint silhouette of us ghosts over the view. The train pulls away with a gentle lurch. Beat 4 — The shimmering sea (14–20s): Through the window, the town gives way to open coastline. Sunlight scatters across the water in thousands of moving sparkles — real specular glitter, not CGI bloom. Utility poles and wires strobe past in the foreground. We rest our chin lightly (subtle downward frame tilt), just watching. Beat 5 — Rails into the sea (20–25s): The track curves and runs so close to the shore that the rails seem to vanish straight into the water — the sea fills nearly the whole frame, waves lapping right up to the ballast. Foam catches the light. We're gliding along the edge of the ocean, the horizon line steady, gulls crossing in the distance. Hold on this until the shot ends. Technical spec: continuous POV, first-person, no cuts or transitions, natural handheld micro-motion, realistic morning ambient audio (station hum, cat, train chime, rail clatter, distant surf), authentic Japanese coastal-railway setting (Shimonada / Enoden-style seaside line), overcast-to-clear soft daylight, photoreal skin and fabric texture, 2.39:1 or 16:9, ultra-realistic.

视频时长：10 秒 画面比例：16:9 核心主题：田园背景下三只幼犬的幽默 “演戏” 挑战 分镜 1：[0-1 秒] 镜头指令： 中特写 (Medium Close-up)，水平固定视角。 画面描述： 背景是色彩饱和度极高的郁郁葱葱的玉米地，叶片纹理清晰可见。两只胖乎乎的幼犬并排坐在干燥的泥地上。一只人类的手突然入镜，对准左侧的花白小狗比出 “手枪” 手势。 动作 / 视觉效果： 听到模拟枪声后，花白小狗配合地向后翻倒，展示出毛茸茸的白色肚皮，动作憨态可掬，伴随轻微的泥土扬尘。 分镜 2：[1-3 秒] 镜头指令： 中特写 (Medium Close-up)，镜头平滑向右横移 (Pan Right)。 画面描述： 手势移向中间的浅黄色小狗。 动作 / 视觉效果： 随着手势发出指令，浅黄色小狗闭上眼睛，顺势向后仰倒。此时，镜头右侧露出了第三只黑色幼犬（小黑），它紧闭双眼，表情严肃且纹丝不动，呈现出一种 “拒不配合” 的滑稽感。 分镜 3：[3-8 秒] 镜头指令： 特写 (Close-up)，固定镜头聚焦于 “小黑”。 画面描述： “小黑” 处于画面中心，它的毛发在自然的阴天柔光下泛着微光。它原本紧闭双眼，倔强地保持坐姿。 动作 / 视觉效果： 当它听到旁白提到的 “鸡腿威胁” 时，瞳孔瞬间发生微细的收缩并猛地睁大眼睛，眼神从迷茫瞬间转化为充满渴望的惊讶，这是表演转折的核心点，表情极具拟人化张力。 分镜 4：[8-10 秒] 镜头指令： 特写 (Close-up)，固定视角。 画面描述： “小黑” 为了美食决定展现极致演技。 动作 / 视觉效果： 手势再次发出 “Biū” 的指令，黑色小狗张大嘴巴做出夸张的惨叫状，全身肌肉紧绷，以一种充满戏剧性的张力向后翻滚。动作与前两只狗形成鲜明对比，充满了表演的热情与喜剧冲突。 场景风格总结： 视觉纹理： 极致细腻的幼犬毛发质感，背景玉米地的绿色要清新明亮，与泥土地的土褐色形成对比。 灯光： 自然明亮的户外柔光，模拟阴天环境，避免生硬的阴影，确保聚焦于小狗的面部微动作。 氛围： 幽默诙谐的纪实风格，强调宠物与人之间的戏剧互动

【整体风格】 电影级古风喜剧，将真实实拍的传统戏班后台与一名纯 Flat 2D Q 版人类贴纸角色融合，围绕 Seedance 2.0 的参考素材引导表演、可控灯光与阴影、稳定物理交互、同步音效和多镜头调度进行设计；采用克制的武侠美学、暖琥珀色灯笼光与清冷后台暗部的对照、层叠纱幕、图形化剪影、经过设计的视觉误导，并以“小人物主动导演自己的英雄形象”构建喜剧弧线，跳脱捣乱与受罚模式。 【角色设定】 Subject ID｜Q版女性小武生 一名 19-23 岁的 Q 版东亚女性小武生，圆润大头约占全身高度的 57%，浅桃色平涂肤色、深棕色豆豆眼、细小自信的嘴巴，黑发盘成高位环形发髻，系红色流苏发带，从脚到头穿白色布靴、黑色绑腿、金边大袖红色织锦戏袍、窄玉腰带，并持一柄迷你直剑。 全程只保持赛璐璐平涂、粗墨线描边与僵直贴纸景深。 【镜头 1｜0-5s｜全景】 锁定机位配合缓慢推轨，建立一个真实传统戏班后台。 画面包含深色木梁、悬挂的红色戏服、青铜挂钩、层叠乐器箱、半透明白色纱幕、一盏落地灯笼、缓慢漂浮的香雾、散落的化妆刷与磨损木地板。 她的完整身体站在灯笼旁，而她的巨大影子覆盖整面纱幕。 【镜头 2｜5-10s｜中景偏牛仔景】 同一个穿红色织锦戏袍的 Q 版女性小武生、同一柄迷你直剑、同一盏灯笼与同一面纱幕，background remains the same real opera backstage。 镜头进行横移跟拍。 同一个穿红色织锦戏袍的 Q 版女性小武生只完成一套干净的三拍剑式——踏步、转身、横斩。 与此同时，一只皮肤纹理真实的真人手轻轻把灯笼向前推，使她的影子扩张成顶天立地的女侠剪影，巨大衣袖与长剑严格同步她的真实动作。 衣料掠风声、木地板脚步声、剑刃破风声和一记低沉戏曲鼓点精准卡点。 【镜头 3｜10-15s｜特写】 同一个穿红色织锦戏袍的 Q 版女性小武生，background remains the same lantern-lit backstage。 真人手意外把灯笼拉远，她威风凛凛的巨大影子骤然缩成一个小点。 她停顿一拍，眯起豆豆眼，用迷你直剑勾住灯笼提环，把灯笼重新拖回准确位置，恢复顶天立地的英雄剪影。 随后，她保持庄严剑式，却向镜头投来一个得意的侧眼。 结尾定格在她微小平面身体与巨大影子形成的强烈尺度反差上。 【技术要求】 9:16 竖屏，使用最高可用原生分辨率，电影级 8K 观感细节，使用一张锁定角色参考图、一张真人场景参考图与一段戏曲鼓点音频参考。 首镜之后加入轻微手持呼吸感，服装与道具严格连续，灯笼移动和影子缩放符合物理逻辑，无对白，无可识别文字，不出现动物，不做拟物角色。 【负面词】 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

【风格】 真实工业液压机实拍场景与扁平2D动漫贴纸角色结合的反差喜剧短片。 工业设备、金属工作台、车间背景保持真实摄影质感；水豚噜噜始终保持纯粹的扁平2D动漫贴纸风格：平面卡通色块、清晰稳定的深色描边、轻微纸片质感、简洁二维阴影和明显贴纸感。 角色只有轻微脚底接地阴影，不产生三维实体质感。 【时长／画幅】 视频时长约14秒，9:16竖屏。 全程固定机位、连续单镜头、一镜到底，无剪辑、无转场。 【场景】 真实工业车间内部。 画面中央下方是一座圆形厚重金属工作台，画面正上方是一台大型液压机的圆形金属压头。压头、角色和工作台位于同一条垂直中心轴线上Mess 工作台表面存在真实金属划痕、油渍和冷硬反光。背景可见液压机框架、管线和工业设备结构，环境照明为冷白色车间灯光。 水豚噜噜从第一帧开始站在圆形工作台中央。 【角色】 主角为水豚噜噜（Capybara Lulu）。 Q版拟人卡通水豚，淡黄色圆润身体，大头短身比例，短小圆润的双臂和双腿，豆状手爪和脚爪，穿橙色短裤和黄色围兜式小围裙，头顶固定一颗小橘子。 角色全程保持相同的五官、身体颜色、服装、描边、头身比例和二维贴纸质感。 头顶小橘子、橙色短裤和黄色围裙始终与身体连接，压扁时跟随身体产生二维挤压形变，恢复后回到原本结构和位置。 【核心道具】 大型工业液压机。 圆形金属压头只能沿垂直方向稳定下降和上升。压头首先接触水豚噜噜头顶的小橘子，然后继续压到头部、身体和四肢。 人物形变必须由压头真实接触触发。压头尚未接触角色时，水豚噜噜保持正常高度和比例。 【剪辑手法】 整条视频只有一个固定正面镜头。 摄影机不移动、不推拉、不变焦、不摇镜。所有画面变化仅来自液压压头上下运动、水豚噜噜的表情动作以及身体压扁和恢复。 结尾不切镜，通过“完全压扁后的停顿—突然恢复—奶凶回应”形成反转笑点。 【镜头1】 水豚噜噜站在金属工作台中央，身体自然放松，左右轻微晃动，嘴巴微微开合，发出软萌的： “嗯呜～唔唔～” 它先看向前方，随后听见上方传来轻微机械声，抬起头观察液压压头，表情从呆萌逐渐变成疑惑。 它歪头发出短促声音： “嗯？唔？” 【镜头2】 液压机启动，压头沿垂直方向稳定向下移动。 水豚噜噜抬头盯着逐渐靠近的压头，眼睛睁大，双臂紧张地抬到胸前，双脚仍站在工作台中央。 它连续发出慌张的小动物叫声： “呜？呜呜！” 压头未接触身体前，角色保持完整正常比例。 【镜头3】 压头首先碰到水豚噜噜头顶的小橘子。 接触瞬间，小橘子、头部和身体开始产生连续的二维挤压形变。随着压头持续下降，角色纵向高度逐渐缩短，头部变矮，脸部五官的上下距离缩短，身体和四肢逐渐被压低，整体轮廓向左右铺开Mess 水豚噜噜睁大眼睛，嘴巴张开，发出逐渐拉长的可爱惊叫： “呜哇——！呜呜呜——！” 声音随着身体被压低逐渐变闷，但角色脸部始终位于身体中央并保持清晰可识别。 【镜头4】 压头继续稳定下降。 水豚噜噜被逐层压成低矮、宽扁的贴纸形态。头顶小橘子变成横向压扁形状，耳朵、脸颊、双臂、身体、橙色短裤和黄色围裙一起向左右铺开。 角色底部始终紧贴金属工作台，不能悬空。 压头到达最低位置后，水豚噜噜被完全压成一张摊在工作台上的扁平二维贴纸。脸仍位于中央，眼睛委屈地眯起，嘴巴保持可见。 接触区域短促爆出少量橙黄色金属火花，同时出现清晰的金属冲击声。 【镜头5】 液压压头快速向上回升。 压头离开后，水豚噜噜仍然保持完全压扁的薄片状态，安静停留片刻，没有马上恢复。 它从压扁的脸部发出微弱、闷闷的： “唔……呜……” 随后眼睛慢慢变成不爽的半眯状态，眉眼向内压，表情从委屈转为奶凶生气。 【镜头6】 水豚噜噜进行快速两段式恢复Mess 第一阶段：压扁身体的中央突然向上鼓起，形成极短暂的矮胖中间形态，左右铺开的轮廓同时向中心收拢。 第二阶段：头部、身体和双腿极速向上拉伸，瞬间恢复成最初的完整Q版站立比例，脚底仍落在原来的工作台中央位置。 恢复瞬间发出短促、弹性的： “啵！” 恢复后，水豚噜噜立即双手叉腰，双脚分开站稳，身体略微前倾，半眯眼、鼓起脸颊，面对镜头和操作者连续发出奶凶的小动物抗议声： “哼唔！呜呜！哼！” 嘴巴随叫声连续开合，声音清楚有力。最后停在双手叉腰、生气又可爱的回应姿势。 【环境声与音效】 全片无背景音乐。 保留真实工业车间环境底噪、液压设备启动声、持续下压的机械轰鸣、金属摩擦和挤压声、最低位置的短促冲击声、少量火花爆裂声以及压头回升时的液压释放声。 水豚噜噜的所有声音均来自角色本人，并与嘴型、表情和身体状态同步。 压扁时声音逐渐变闷；恢复瞬间加入短促的卡通弹性回弹音；恢复后奶凶叫声重新变得清晰响亮。

【基础设定】 生成一段约 10秒、9:16竖屏 的反差搞笑短视频。 真实住宅小区实拍环境，与Q版扁平2D动漫贴纸角色合成。 全程固定鱼眼机位，一镜到底，无剪辑、无转场、无黑场、无推拉摇移、无旋转、无变焦。 不添加背景音乐、字幕、文字、门禁UI、Logo或水印。 【核心剧情】 水豚噜噜从远处跑到门禁立柱旁，听到门禁系统提示“请刷脸”后，认真抱住立柱，把脸凑到鱼眼摄像头前。 第一次识别失败后，它向后收回，疑惑地眨眼，再次更加用力地贴近摄像头，但系统仍然提示“未识别”。 整体表演呆萌、认真、自然，形成反差喜剧效果。 【场景】 真实现代住宅小区或社区入口，白天自然阳光。 画面包含： 浅色住宅建筑； 绿色树木和灌木； 灰色铺装道路； 地面斑驳树影； 画面右侧固定的绿色门禁立柱。 场景中不得出现居民、保安、车辆、其他动物或无关角色。 【摄影机与鱼眼效果】 摄影机模拟固定安装在绿色门禁立柱上方的刷脸摄像头，摄像头本体不入镜。 保持明显、自然的超广角鱼眼桶形畸变： 建筑和道路边缘自然弯曲； 画面两侧产生拉伸； 绿色立柱具有近距离透视； 边缘带轻微暗角； 角色靠近镜头时快速夸张放大。 角色脸部贴近摄像头时，占据画面约 70%至85%。 脸部放大只能由角色真实向摄像头靠近产生，禁止摄影机推进、数字变焦或直接缩放角色贴图。 【角色】 唯一角色为： 水豚噜噜（Capybara Lulu） Q版拟人卡通水豚，具有： 淡黄色圆润身体； 大头短身的Q版比例； 短小圆润的双臂和双腿； 豆状手爪和脚爪； 橙色短裤； 黄色围兜式小围裙； 头顶固定一个小橘子。 角色全程保持相同的五官、身体颜色、服装、描边和头身比例。 头顶小橘子、橙色短裤和黄色围裙不得掉落、消失或变形。 【角色视觉风格】 水豚噜噜始终保持纯粹的 扁平2D动漫贴纸风格： 平面卡通色块； 清晰稳定的深色描边； 轻微纸片质感； 简洁二维阴影； 明显的扁平贴纸感。 可以保留轻微脚底接地阴影，但不得产生3D实体感。 禁止生成真实水豚、半写实动物、毛绒玩偶、软胶公仔、塑料模型或三维CG角色。 禁止真实毛发、毛绒纤维、塑料反光、橡胶材质、立体皮肤纹理和三维体积光影。 【构图与运动路线】 开头为固定鱼眼大全景。 水豚噜噜从画面中远景偏右区域出现，位于绿色门禁立柱左后方，角色体积较小，住宅小区环境占据主要画面。 完整运动路线： 中远景出现 → 快速小步跑向摄像头 → 停在绿色立柱左侧 → 抬头看向摄像头 → 听到“请刷脸” → 靠近并抱住立柱 → 第一次贴脸 → 听到“未识别” → 向后收回 → 第二次贴脸 → 再次听到“未识别”。 【动作时间轴】 【开头：跑向门禁】 水豚噜噜从中远景偏右位置，迈着连续、快速、短小且略带弹跳感的步伐跑向摄像头。 双腿自然交替迈步，双脚清楚接触地面，双臂随步伐前后摆动，身体有轻微上下弹动。 角色随着接近摄像头自然变大，禁止漂浮、滑行、瞬移、跳帧或直接放大贴图。 【停下：等待刷脸】 水豚噜噜自然减速，停在绿色门禁立柱左侧，双脚稳定站在地面上。 它抬头看向立柱上方的摄像头，嘴巴微微张开，表情从开心变成疑惑和等待。 保留自然眨眼、轻微呼吸和重心变化。 门禁系统发出清楚、冷静的中文机械语音： “请刷脸。” 水豚噜噜停顿片刻，认真理解系统提示。 【抱柱：靠近摄像头】 水豚噜噜转向身旁的绿色立柱，迈近一步。 动作顺序必须为： 单手接触立柱 → 另一只手抱上 → 双臂环抱立柱 → 身体贴近立柱左侧 → 利用双臂拉力，将头部和上半身向摄像头前探。 双手位于立柱前方，部分身体位于立柱左后方，立柱自然遮挡部分躯干。 手爪必须与立柱表面真实接触，身体不得穿过立柱。 角色不能攀爬、跳起、悬浮或飞向镜头，双脚始终接触地面。 【第一次贴脸失败】 水豚噜噜保持双臂抱住立柱，将脸逐渐凑近鱼眼摄像头。 脸部被鱼眼透视自然放大，占据画面约70%至85%。 它正面认真盯着镜头，双眼自然睁大，嘴巴微微张开，表情呆萌、认真而茫然。 门禁系统发出机械语音： “未识别。” 【收回并再次尝试】 听到“未识别”后，水豚噜噜仍然抱住立柱，将头部和上半身自然向后收回一段距离。 双手不松开，双脚基本不移动，脸部随着真实后退自然缩小。 它疑惑地眨眼，短暂停顿后，认为自己距离还不够近，再次更加果断地将脸凑向摄像头。 【第二次贴脸失败】 水豚噜噜再次贴近鱼眼摄像头，脸部重新夸张放大并占据画面主体。 角色继续保持双臂抱柱、双脚接地，正面盯着镜头，神情认真、呆萌、茫然。 门禁系统再次发出机械语音： “未识别。” 语音结束后自然停留片刻。 最后一帧保持水豚噜噜仍然抱住绿色立柱、脸贴近摄像头的状态。 【声音】 不添加背景音乐。 保留： 住宅小区白天环境底噪； 树叶和远处自然环境声； 水豚噜噜跑动的轻快脚步声； 手爪接触、抱住立柱的轻微摩擦声； 门禁电子提示音； 清楚的中文机械系统语音。 系统语音顺序： 停下抬头后：“请刷脸。” 第一次贴脸后：“未识别。” 第二次贴脸后：“未识别。” 禁止角色对白、旁白、观众笑声、额外系统语音和无关音效。 【强制约束】 全片一镜到底，只生成一次完整连续动作。 摄影机全程固定，保持明显鱼眼畸变。 水豚噜噜始终为扁平2D动漫贴纸角色。 角色只能有两条手臂和两条腿，不得出现额外肢体或四足爬行。 跑动具有连续脚步和稳定接地关系。 抱柱及贴脸阶段双脚始终接触地面。 两次贴脸动作必须完整出现，中间必须有明显的向后收回。 三句系统语音必须与对应动作准确同步。 手臂、身体和立柱保持正确遮挡，禁止穿模。 绿色立柱不得移动、弯曲、软化、断裂或变形。 极近距离时，角色五官、脸部轮廓和头顶小橘子保持清晰稳定。 不新增人物、动物、车辆、文字、界面或无关剧情。 【负向提示词】 真实水豚，写实动物，半写实角色，真实毛发，毛绒玩具，布偶，软胶公仔，塑料模型，3D角色，三维CG，体积光，塑料反光，橡胶材质，五官融化，五官错位，角色变形，身份变化，多余肢体，肢体缺失，服装变化，橘子掉落，角色穿模，立柱变形，立柱移动，四足爬行，攀爬，跳起，悬浮，飞行，漂移，滑行，瞬移，跳帧，脚底离地，直接缩放角色，摄影机运动，推进，后拉，摇镜，跟拍，旋转，变焦，弱鱼眼效果，摄像头入镜，黑场，转场，重复镜头，背景音乐，旁白，角色对白，笑声，字幕，文字，乱码，门禁UI，水印，Logo，其他人物，其他动物，车辆。

15秒、16:9。平成時代の真夏の日本の田舎を舞台にした、フォトリアルな実写映像と、輪郭線だけのチョークアニメーションを融合したVFX動画。最初から最後までカットを切らない、少年視点の完全なワンカット。 【実写空間】 真夏の昼下がり。古い木造一戸建ての畳敷きの和室。色あせた襖、障子、乾いた木の柱、ちゃぶ台、蚊取り線香、ブラウン管テレビ、古い扇風機、壁掛けエアコン。奥の台所には古い白い冷蔵庫がある。 障子越しの強い夏の日差し、湿った暑い空気、わずかな陽炎。1990年代の家庭用フィルムカメラで撮影したような、褪せた夏色、柔らかなフィルム粒子、軽い色にじみ、わずかな白飛びとフレームの揺れ。 【カメラ】 カメラは少年の目の高さにある完全なPOV。少年が手持ちカメラを持って和室から台所へ歩く。自然な手ブレ、呼吸に伴う上下動、アニメーションを少し遅れて追うパンと前進。 0〜10秒は少年の手、腕、顔、身体を一切映さない。少年の右手が映るのは、10秒以降に冷蔵庫を開ける一度だけ。 【手描きアニメーションの絶対的な画風】 アザラシ、ペンギン、シロクマは、すべて粗いチョークの輪郭線だけで描く。身体の内側は完全に透明な空洞であり、実写の畳、壁、家具、冷蔵庫が動物の身体を通して常に見える。 動物を白や青で塗りつぶさない。不透明なシルエットにしない。身体内部に色、陰影、毛、立体的な肉付けを描かない。輪郭線、小さな点の目、短い口の線だけで動物を表現する。 線は均一なデジタル線ではなく、太さが揺れ、途切れ、粉が落ち、毎フレーム描き直される。白、水色、濃い青のチョーク線を使い、一部に黄色とピンクの小さなアクセント線を入れる。 賑やかさは動物を塗ることで作らず、周囲を飛び交う大量の小さな線画で作る。雪の結晶、泡、渦巻き、魚の輪郭、氷の欠片、星、波線、冷風のリボン、スピード線が、異なる大きさと速さで前景・中景・背景を動く。画面全体に楽しい動きがあるが、実写の部屋は見失わない。 【0〜4秒：扇風機とアザラシ】 畳の上で古い扇風機が首を振っている。羽根から白と水色のチョーク線による冷風の渦が次々と飛び出す。 複数の渦線が空中でつながり、輪郭線だけの大きなアザラシへ連続的に変形する。アザラシの身体内部は完全に透明で、背後の畳と襖が透けて見える。 アザラシは楽しそうに空中を泳ぎ、身体をくねらせてカメラの周囲を一周する。周囲には小さな魚の輪郭、泡、星、氷の粒、波線が群れになってついてくる。アザラシの尾から何本ものチョーク線の残像が伸びる。カメラは少し遅れてアザラシを追い、壁のエアコンへ向かって大きくパンする。 【4〜8秒：エアコンとペンギン】 アザラシがエアコンの前を横切ると、吹き出し口から白と青のチョーク線による冷気が勢いよく噴き出す。 冷気の線が5羽の輪郭線だけのペンギンへ滑らかに変形する。すべてのペンギンの身体内部は透明で、背景の壁や家具が透けて見える。 5羽は同じ大きさで並ばず、手前に大きな1羽、中景に3羽、奥に小さな1羽を配置する。それぞれが異なる高さと速度で、海中を泳ぐように和室を飛ぶ。1羽は回転し、2羽は交差し、1羽はちゃぶ台の下をくぐり、1羽はカメラのすぐ横を素早く通過する。 ペンギンの周囲には、雪の結晶、泡、渦、魚の輪郭、リボン状の冷風、細かなスピード線が大量に流れる。風鈴、暖簾、蚊取り線香の煙が冷風で揺れる。カメラはペンギンたちを追って襖を抜け、奥の台所へ進む。 【8〜10秒：冷蔵庫への移動】 5羽のペンギンとアザラシが台所へ入り、古い白い冷蔵庫の周囲を旋回する。動物たちの透明な輪郭線が、冷蔵庫の手前と奥を交互に横切り、正しい遮蔽と奥行きを作る。 冷蔵庫の取っ手へ向かって、青いチョークの矢印、雪の結晶、魚の輪郭が集まる。カメラは冷蔵庫の正面で止まる。この時点まで少年の手や身体は一切映らない。 【10〜12秒：少年が冷蔵庫を開ける】 ここで初めて、画面下から実写の日焼けした少年の細い右腕と手だけが入る。手には5本の自然な指があり、古い冷蔵庫の取っ手を正しく握る。 少年が扉を手前へ引き開ける。扉の重さに合わせてPOVカメラも少し傾く。少年の顔、左手、胴体、脚は映らない。冷蔵庫は自動で開かない。 【12〜15秒：シロクマとかき氷】 冷蔵庫の青白い実写の光が台所と少年の手を照らす。 冷蔵庫の中には、輪郭線だけで描かれた丸いシロクマが座っている。シロクマの身体内部は完全に透明で、背後の冷蔵庫の棚や食品が身体を通して見える。 シロクマだけを塗りつぶさない。一方、目の前のガラス器に入ったかき氷は、鮮やかな水色と青の細かなチョーク線を何層も重ね、きらきらした密度のある描写にする。透明なシロクマと、賑やかなかき氷の対比を作る。 最初、シロクマは顔を下に向け、スプーンでかき氷を食べている。冷蔵庫が開く音でスプーンが途中で止まる。一瞬静止した後、顔をゆっくり上げ、最後にカメラとまっすぐ目を合わせる。口元には水色の短いチョーク線による、かき氷の跡が少し付いている。 その瞬間、アザラシと5羽のペンギンが冷蔵庫の周囲から一斉に顔をのぞかせる。雪の結晶、魚、泡、星、渦巻き、冷風のリボンが画面いっぱいに弾ける。ただし、すべて輪郭線と細い落書き線であり、動物の身体は透明なまま。 少年の手は取っ手を握ったまま残る。シロクマと少年の視線が一致すると、冷たい白い空気がレンズへ流れ、画面の縁が薄く曇って終了する。 【音声】 BGM、ナレーション、会話なし。遠くの蝉、古い扇風機のモーター、風鈴、畳の足音、衣擦れ、エアコンの動作音、冷蔵庫の扉、冷蔵庫内部の小さな機械音。 手描きアニメーションには、チョークが紙を走る軽い音、冷風、泡、水しぶき、小さな氷が触れ合う涼しい音を同期させる。最後はシロクマのスプーンとガラス器が触れる小さな音を強調する。 【厳守事項】 最後までワンカットと同じPOVを維持する。動物の身体内部は全フレームで完全に透明にし、白、青、黒などで塗りつぶさない。不透明なシルエット、セル画のベタ塗り、実写動物、3DCG動物、立体的な毛並みへ変化させない。 0〜10秒は少年の手や身体を絶対に映さない。冷蔵庫は10秒以降に少年の右手で開ける。シロクマは最初からカメラを見ず、かき氷を見た状態からゆっくり顔を上げる。 カット、場面転換、瞬間移動、字幕、文字、ロゴ、透かし、BGMを追加しない。

【整体风格】 大师级电影写实风格结合精致古风时装画册美学，写实真人，严格参考上传图片，场景为浅灰白无缝无影棚，背景与地面自然连成一体，采用柔和通透的前上方漫射光，保留细腻肤质与轻盈半透明面料质感，镜头为 9:16 竖屏固定高机位前方俯拍，向下约 30–35 度。 【主体设定】 一位 20–24 岁的东亚年轻女子，身形修长匀称、气质清冷温婉，鹅蛋脸，冷白肤色，大杏眼，平直眉，小巧鼻尖，柔和粉唇，黑色古风长发，半盘发造型，银色花簪头饰，珍珠垂坠耳饰。 初始造型为同一位女子身穿白色外披大袖袍、蓝色绣纹交领内层、浅蓝色长裙，腰间银色腰饰与流苏，裙摆向画面下方展开。 【固定贴纸设定】 画面四周全程固定悬浮四个完整人物贴纸，均为真实身体轮廓抠像并带清晰白色虚线描边，绝不是矩形缩略图： 左上为白色上衣 + 浅蓝色百褶长裙 + 蓝色腰带 右上为白色内搭 + 蓝色无袖马甲式长裙并带银色绣纹 左侧中部为浅蓝色内裙 + 深蓝色交领层次 + 白色外披长袖袍 左下为浅灰紫色外披 + 米白色内搭 + 深蓝色绣纹下裙 【镜头段落 1｜0.00–2.95 秒】 同一位女子穿初始造型，从画面下方偏右轻缓靠近镜头，身体微微前倾，宽大袖摆与裙摆自然轻摆，抬眼看向高机位镜头； 一个白色手形鼠标指针点击左上贴纸，复制出一个相同的完整人物贴纸，从左上斜向飞向中央并迅速放大，放大过程中白色虚线描边始终紧贴人物轮廓，约 1.25 秒时与头部、肩膀、腰部和动作姿势完全重合，中央同一位女子无缝换装成左上白衣浅蓝长裙造型并继续原来的靠近动作； 随后她一只手自然落在腰前，另一只手轻提袖口，带淡淡微笑，向左缓慢转身，逐渐露出侧面与半背面，发丝与头饰轻轻摆动。 【镜头段落 2｜2.95–5.40 秒】 白色鼠标点击右上贴纸，复制出的完整人物贴纸从右上斜向飞入中央，约 2.95 秒时同一位女子切换成白色内搭 + 蓝色无袖马甲式长裙，背景始终保持同一浅灰白无影棚，四个原始贴纸人物也始终固定保留； 同一位女子继续同一个转身方向，随后转回正面，衣摆与发丝自然回落，身体短促下沉，双臂轻微向两侧展开，让袖摆自然打开，然后鼠标点击左侧中部贴纸，该贴纸复制体从左侧飞向中央，约 4.40 秒时同一位女子无缝换装成浅蓝内裙 + 深蓝交领层次 + 白色外披长袖袍，脸部、发型、头饰、耳饰和身材比例完全一致。 【镜头段落 3｜5.40–8.00 秒】 同一位女子双手收于身前，低头轻摆肩膀，连续左右自然歪头，随后抬眼看向镜头，再将视线转向左下贴纸人物； 鼠标点击左下贴纸，复制出的完整人物抠像从左下斜向右上飞入中央并快速放大，约 6.55 秒时与中央人物完美重合，完成第四次换装，变为浅灰紫外披 + 米白内搭 + 深蓝绣纹下裙； 接着她一只手轻扶衣襟，另一只手自然垂落，先看向侧面，再转回正面，自然歪头，抬眼直视高处镜头，以优雅、清冷、强古风感的姿态结束，长袖与裙摆在结尾轻微自然垂落。 【换装时间点】 四次换装卡点严格落在： 1.25 秒 2.95 秒 4.40 秒 6.55 秒 【技术要求】 全程固定机位，不推拉、不摇镜、不俯仰、不旋转、不变焦； 换装必须仅由完整人物贴纸飞入、放大、重合实现，不能出现单独衣服飞入、人物残影、双重人物、服装叠穿； 声音仅保留鼠标点击声、贴纸飞入时的短促移动音效、换装完成时清晰利落的卡点音效，以及轻微衣料摩擦声，不要背景音乐。 【负面词】 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

参考视频：video ref.mp4 角色设计参考：@mayz.png 人物身份参考：@image1.png 生成一段 7秒、9:16竖屏的仙侠时尚视频，全程采用同一个连续镜头，一镜到底。 角色的完整服装、发型、头饰、首饰、鞋子、身材比例与整体造型严格参考 @mayz.png 并保持一致；人物脸型、五官、身份特征与识别度严格参考 @image1.png。即使拍摄角度和画面方向发生变化，也必须始终明确为同一个人物。不得改变服装结构、颜色、材质、发型或配饰位置。 整条视频的摄影机运动路径、人物动作节奏、迈步速度、镜头旋转时机、旋转方向、构图变化以及结尾姿势，严格参考 video ref.mp4 完整复刻，不得自行修改。 场景设置在清晨云海之上的白玉仙台。角色位于高山之巅的开阔白玉平台，地面为浅灰白色石材，带有柔和克制的反光。平台外是翻涌云海、青灰色远山与明亮清透的青蓝天空，远处可见轮廓简洁的白色仙门。白玉平台边缘与远山地平线必须清晰可见，作为摄影机旋转时稳定的空间参照。 清晨阳光从人物侧后方照来，在发丝、刺绣、透明披袖与层叠薄纱边缘形成柔和的浅金色轮廓光。贴近镜头的前景只保留少量严重失焦的白玉构件、青铜装饰边缘与零散浅色花瓣，不出现草地、密林或杂乱植物，也不能遮挡人物鞋子与裙摆。 镜头开始时，摄影机几乎贴近白玉地面，使用极低机位超广角镜头。靠近镜头的一只银色高跟凉鞋被透视夸张放大，占据画面前景。角色以自然速度从镜头上方迈步经过，她的双腿、层叠薄纱裙摆底部与飘动的披袖占据画面大部分区域。 摄影机随后平滑向上升起，镜头持续保持仰拍，并沿着角色的运动轨迹抬镜跟随，从鞋子、腿部和裙摆逐渐移动至她的腰部、上半身和面部。裙摆、披袖与发饰在自然移动和晨风中产生轻盈连续的飘动，服装结构必须始终稳定，不得变形或突然变化。 当角色面部在大约 1.8秒 时进入画面，摄影机在同一个连续镜头中突然快速进行一次顺时针90度光轴旋转，形成强烈、迅猛的“雷霆甩镜”效果，并伴随短暂、自然的方向性运动模糊。 旋转过程中，人物、白玉平台边缘、云海、远山、天空、仙门与全部环境必须保持同一空间关系，并跟随摄影机同步连续旋转，直到整个画面完全形成横向侧转构图。禁止只旋转人物、只旋转背景、跳过中间角度或直接切换到旋转后的画面。 摄影机旋转完成后，角色严格保持参考视频中的最终身体姿态、手部位置、眼神状态与停留节奏。人物表情自然、冷静、克制，带有清透仙气与高级时尚感，不增加额外动作。 全程禁止剪切镜头、硬切、隐藏切镜、闪白、黑帧、画面重置或更换机位。禁止换装、人物变脸、肢体变形、服装错乱、背景跳变、旋转方向错误或结尾姿势改变。 不添加背景音乐，只保留晨风声、轻微衣料与薄纱摩擦声、高空环境声以及鞋子踏过白玉地面的真实声音。

15秒，实拍电影质感，暗黑电影感地府权力组织主题，全程真正一镜到底，无剪辑、无跳切、无转场、无隐形遮挡剪辑。无对白，无背景音乐，无环境声，无任何音效。 场景位于一处极简、庄严、肃穆的阴司审判空间。整体空间高挑、空旷、克制，像古老中式地府审判厅与现代极简权力空间融合而成。空间采用深灰黑色石质地面，地面带轻微反光；墙面为冷灰色巨大石壁与暗色中式立柱结构，局部有极简黑色金属框架与竖向门廊。远处可见一扇模糊而巨大的惨白冥门光源，像幽冥之门。整体光线偏暗，但人物必须清晰可见：冷白主光从顶部与侧前方斜射进入，形成克制的几何光区；少量暗红色环境反光轻微映在地面与衣摆下缘；空间内有极轻微雾气与体积光，增强层次感。整体视觉高级、冷峻、极简、神秘，具有东方暗黑审判电影的气质。 五名人物从拍摄开始便分散站在同一个连续空间内，彼此保持约1.5至2米距离，形成松散但有层级暗示的分布。人物位置从一开始就固定存在，不能突然出现、瞬移或改变方位。镜头通过推进、绕行、横移和人物身体前景逐步揭示他们，但不能穿过人物、衣摆、武器、墙壁或空间结构。 人物身份与外形始终严格保持参考图一致： 白无常@参考图1：白色高腰短裤，白色新中式高定结构感上衣，半透明魂幡式披帛，手持“招魂”令牌，深栗棕色短发，烈焰红唇，冷艳诡美。 黑无常@参考图2：黑色修身长袍，分层开衩垂裳，丰腴身材，长卷发低双马尾盘发，手持“引路”令牌，冷酷压场。 牛头@参考图3：正常人脸，粗壮牛角，中式重装镇狱武官穿搭，手持粗重锁链，年轻、魁梧、沉重。 马面@参考图4：正常人脸，马耳，中央竖立长发束，中式追魂猎手穿搭，手持拘魂索，修长、锐利、阴冷。 阎罗王@参考图5：42岁，172cm，身材挺拔但不健壮，黑红色中式审判长袍，黑金腰饰，玉戒，手持“生死簿”，压轴登场，绝对裁决者。 【0.0—3秒｜白无常极近特写】 画面从45度倾斜构图开始。前景是一根距离镜头极近的白无常食指，指尖虚化，占据画面一角，背景是白无常中景照。镜头从她的手指前景推进，直到胸部，然后上升推到白无常的完整面部特写。她的脸必须完整清晰，鹅蛋脸、浅棕瞳孔、圆润杏眼、烈焰红唇清楚可见。镜头在她整张脸上稳定停留约0.8秒，确保面容完整清晰。 随后镜头扫过她的左肩顺肩膀而下再拉远至人物中远景，白无常轻轻抬起下巴，一只手中的令牌显现，身体保持原地不动。镜头要清晰看到她的白色高腰短裤、肩线、胸线、腰线，以及令牌上的 “招魂” 二字。 【3—5秒｜绕肩揭示黑无常】 镜头从白无常正面开始向画面右侧平滑横移，同时以顺时针方向绕过她的左肩。白无常的肩膀和白色披帛短暂成为自然前景，但不能完全遮住画面。 横移初段速度较慢，绕过肩膀后突然加速，迅速揭示站在她右后方约1.8米处的黑无常。镜头快速推进聚焦到黑无常的面部特写，并显现她丰腴但克制的轮廓、黑色修身长袍与分层开衩垂裳。黑无常轻轻触碰一下“引路”令牌，随后令牌自然垂落，目光冷冷追随镜头，微微偏头。令牌上的 “引路” 二字必须清晰可见。 【5—7秒｜快速横移揭示牛头】 镜头向左侧加速横移，但保持稳定，不出现晃动或突然甩镜。镜头在接近左侧区域时迅速减速，平稳停到牛头的面部近景。牛头保持正常年轻男性人脸，牛角清楚可见，冷白皮肤与中式重装皮革护甲清楚可辨。 牛头身体微微低垂重心，随后缓慢抬眼，另一只手中的粗重锁链自然下坠。镜头轻微向他推进，从胸部近景推进至完整面部特写，在他的眼睛与牛角之间稳定停留。镜头还需带到胸甲、牛首腰扣和锁链，强化重型镇压者气场。 【7—9.5秒｜贴肩环绕揭示马面】 牛头保持原地不动，锁链轻微晃动。镜头从他的右侧肩膀旁经过，以弧形路线向后方绕行。镜头移动速度先慢后快，从牛头宽阔的肩线自然带出站在更深处的马面。马面距离牛头约2米，不能与牛头重叠。镜头不能穿过牛头肩膀，也不能从两人距离不足的缝隙强行穿越。 马面的拘魂索首先进入画面前景。镜头沿着拘魂索迅速向上抬升，速度明显加快，最终突然减速，停在马面的完整面部特写。马面是正常年轻男性人脸，头顶马耳与从前额延伸到后脑的竖立中央发束必须清晰可见。马面轻轻侧头，目光锐利地看向镜头，身体微微前倾，手中拘魂索轻轻摆动，不前冲、不夸张挥舞。 【9.5—11.5秒｜低位掠过拘魂索揭示阎罗王】 镜头从马面的身体左侧绕过，在经过他身旁后迅速向前加速，随后转向空间中更深处、靠近冥门光影交界的位置。站在光影边缘的阎罗王被完整揭示。阎罗王一开始以右侧特写侧脸进入画面，原本侧身站立，手持一本黑色书册。然后镜头迅速聚焦到书封面，确保 “生死簿” 三个字清晰可见。 随后镜头从转身至正面的阎罗王推进到他的面部特写，眼球清晰可见，动作沉稳克制。他一只手持生死簿，另一只手可轻触黑金腰饰或玉戒，表情平静、冷酷、不怒自威。 【11.5—15.0秒｜五人汇入最终群像】 镜头从阎罗王面部开始缓慢环绕，同时平稳后撤并略微升高。镜头环绕过程中，之前出现的四人按照其原本物理位置依次重新进入画面，所有人有轻微走动。不能突然滑动、交换位置或从远处高速冲入画面。 最后1.5秒，镜头加速后撤，从多人中景拉成完整全身群像，再缓慢减速并稳定停住。最终站位如下： 阎罗王站最前面正中央，后排左右边缘分别是牛头和马面，后排中间是白无常和黑无常。 五个人站位随意，保持一定距离，但每个人的完整头部、面部、身体与鞋子都清晰可见，彼此不遮挡，道具与武器不遮挡任何人的脸。 五个人同时面向镜头，但神态各不相同： 阎罗王正面直视镜头，冷静裁决； 白无常微微侧脸，嘴角带一丝寒意笑意； 黑无常冷冷斜视镜头，令牌自然垂落； 牛头低头压眉，手握重链，沉重稳固； 马面微微抬下巴，拘魂索下垂，眼神锐利。最终形成一张严肃、神秘、充满等级秩序与压迫感的五人地府审判团合照。镜头完全静止约0.8秒作为结束。 运镜总要求 镜头速度必须呈现明显的忽快忽慢节奏： 极近前景快速推进 → 面部减速停顿 → 平滑绕肩 → 突然加速横移 → 再次减速锁定面部 → 贴肩环绕 → 低位加速滑行 → 阎罗王面部慢速锁定 → 最后快速后撤并逐渐减速定格。 所有加速与减速都要有平滑的速度曲线，不能突然跳帧、瞬移或机械急停。 每个人至少获得一次清晰、无遮挡的面部特写或近景。 全程严格一镜到底，人物形象与服装必须与参考图一致。 场景保持极简、克制、创意感的地府审判空间，不要血腥，不要廉价恐怖片特效，不要过多烟雾遮挡人物。

[CONDITION DEFINITION] 1:1スクエア、15秒、参照画像なし。広大な島全体に港町が高速で組み上がる、シネマティックな高級ミニチュア・ジオラマ形成モンタージュ。方向性は、チープな玩具ではなく、高級な手作りジオラマ、高級コレクタートイ、映画撮影用の精密模型のような質感。可愛らしさ、整理された美しさ、模型ならではの楽しさを残しつつ、島全体が大きく広がる壮大感を出す。明るくクリアな昼光、青い海、白い波、赤・青・白の屋根、木の桟橋、石の岸壁、灯台、橋、運河、丘の上の街並み、複数の港区画を見せる。主役は人物ではなく、巨大な港島ジオラマそのもの。文字、ロゴ、透かし、字幕は不要。 [SHOT/FLOW] 0.0〜3.0秒： 広角の高い俯瞰ショットから開始する。小さな箱庭ではなく、青い海に囲まれた大きな島全体を見せる。何もない島の地形に、海岸線、湾、防波堤、崖、丘、港口、運河、複数の岸壁、橋、主要道路、段差のある街区の輪郭が高速で現れ、まず島と港の大きな骨格が一気に完成する。最初から外海、湾、丘、港の広がりを見せ、こじんまりした印象を避ける。 3.0〜6.5秒： カメラは広角を保ったまま、ゆるやかに下降しながら港側へ前進する。木の桟橋、石畳、階段、手すり、倉庫、漁具小屋、運河沿いの建物、赤・青・白の屋根の家並み、丘へ続く細い道が、カチッとはまるようにテンポよく組み上がる。寄りすぎず、常に港・丘・外海の奥行きが見える構図にする。 6.5〜10.0秒： 自然な場面転換を入れ、港の中距離ショットへ移る。複数の桟橋、小舟、帆船、係留ロープ、木箱、小型クレーン、街灯、旗、橋、運河、岸壁の石の継ぎ目、桟橋の板などの細部が連続して完成する。ここは形成の快感を最も強く見せる場面だが、小物の寄りだけにせず、広い港区画全体が一気に密度を増していくように見せる。 10.0〜12.5秒： 灯台、丘の上の街並み、複数の港区画、外海へ続く湾を見せる。灯台が完成し、屋根、窓、橋、街灯、船、旗が最後の仕上げとして整い、広大な港島ジオラマが完成形に近づく。ここで島全体のスケール感をもう一度強く見せる。 12.5〜15.0秒： 完成した港島に命が宿る。水面がきらめき、小舟が少し揺れ、旗とロープが風になびき、灯台の光がゆっくり回り、窓や街灯が小さく点灯する。最後は広角で大きく引いた壮大な全景。港, 丘, 灯台, 複数の桟橋, 外海まで含めて, 大きな高級ジオラマ島が可愛らしく息づき始めた余韻で締める。 [CAMERA/EDITING] カメラは広角を重視し、島全体の広がりを常に感じさせる。開始は高い俯瞰、次に下降前進、港沿いの近接通過、灯台まわりの軽い回り込み、最後は大きく引いた全景。15秒内に2〜3回の自然な場面転換を入れる。寄りカットは入れてよいが、寄りすぎて小さな玩具だけに見えないようにする。前景・中景・遠景を重ね、港、丘、灯台、外海が同時に見える構図を作る。テンポは速く、形成の快感を重視する。単調な直線移動や長い静止は避ける。 [SOUND] BGMあり、効果音あり。BGMは明るく爽快で、少し冒険感のあるシネマティックな音楽。高級ジオラマが完成していく高揚感を支えるが、派手すぎない。効果音は小さな組み立て音、木や石がはまる音、軽いクリック音、水音、波のさざめき、ロープや旗の音、船のきしみ、遠いカモメ、灯台の小さな機械音。セリフなし、ナレーションなし。 [NEGATIVE] 小さすぎる島、こじんまりした箱庭、安い玩具感、チープなプラスチック感、雑なクラフト感、狭い港、数軒だけの町、スケールのない構図、暗い曇天、濁った海、実物大の巨大人間、人物主役化、文字、ロゴ、透かし、字幕、単調な直線カメラ、完成後に完全静止するだけのラストを避ける。

[Global Setting] Base Environment & Texture: A city rooftop garden at golden hour, warm orange light raking across wet pavement from earlier rain, soft haze in the skyline behind, string lights beginning to flicker on. [Opening beat, 0–8s]: Wide shot. A runner in a grey hoodie jogs into frame along the rooftop path, breath visible in the cool evening air, camera holds static. [Mid beat, 8–22s]: Camera begins a slow tracking pan alongside the runner at hip height, matching pace. Passing shots of string lights, a couple sharing a bench, steam rising from a nearby vent. Light shifts from orange to a deeper blue as the sun dips further. [Closing beat, 22–30s]: Runner slows to a stop at the rooftop edge, camera arcs around to a three-quarter front view, city skyline fully lit behind them, runner catches breath and looks out over the view. Hold on final frame. Audio: Footsteps on wet concrete, distant city ambience, wind, no music.

【风格】武侠电影级动作喜剧（Wuxia Action Comedy），古风客栈实景，8K，达芬奇高级调色，升格慢动作与快剪结合，真实布料与木屑物理，横屏16:9 【时长】30秒 【场景】古风江湖客栈大堂，木桌长凳、红灯笼、蒸笼冒热气，梁柱交错，光从窗棂斜射带浮尘 【角色】女侠A @图1（红衣冷艳，用剑）、女侠B @图2（青衣飒爽，用双筷/软鞭），两人武艺相当 [00:00-00:02] 钩子·同时出手 蒸笼掀开——热气散去，只剩孤零零一个白胖包子。两只手"唰"地同时探向它，在包子上方一寸僵住。镜头拉近两人对视，杀气骤起。 音效：蒸汽"嘶"声，一声兵刃出鞘。 [00:02-00:06] 拔剑·桌面开打 A拔剑挑翻整张桌子腾空，B踩着飞起的长凳跃上房梁，包子被气浪震到空中。碗筷茶壶四散飞溅。 [00:06-00:11] 空中争夺（升格） 包子在半空划弧，两人凌空交手，A剑尖挑、B筷子夹，包子在刀光间被抢来抢去，谁也没碰实，热气还在包子上冒。 [00:11-00:16] 客栈大乱 战场蔓延，两人踩桌翻梁、灯笼被削落、木屑纷飞，其他"食客"抱头躲桌下。包子始终在两人争夺中心悬着不落。 [00:16-00:21] 决胜一击（升格特写） A与B同时使出绝招，剑气与鞭影在包子两侧交汇——包子被劲气托着高高弹起，两人各自摆出接杀造型，屏息等它落下。 [00:21-00:26] 神反转 镜头跟着下落的包子——一只慵懒的橘猫"喵"地从桌上跃起，在半空一口叼住包子，稳稳落地，当着两人的面吧唧吧唧吃完，舔爪。 音效：所有打斗音效戛然而止，一声"喵~"，乌鸦飞过。 [00:26-00:30] 傻眼收尾（定格） A和B保持着帅气的接杀姿势僵在原地，剑和筷子还举着，齐刷刷看向吃饱打嗝的猫，表情裂开。 字幕（砸出）：「有些仗，打赢了也没用。」 音效：一记喜剧"咚"，猫满足地打了个嗝。 【导演备注】1.反差铁律：打斗越是天崩地裂、招式越华丽正经，最后被猫截胡的落差越好笑；2.身份锁定：A=@图1红衣用剑、B=@图2青衣用筷/鞭，脸/服装全程一致；3.节奏：2秒对峙→拔剑→空中争夺→大乱→决胜→猫反转，动作段用快剪+关键处升格；4.物理真实：包子热气、木屑、灯笼、布料惯性、猫的重量都要真实，武侠不卡通；5.包子必须全程在画面中心被争夺，是视觉焦点和笑点核心；6.不见血、不伤人，纯招式对拆。 【负面提示词】卡通武侠，塑料CG，换脸，两人身份/武器混淆，包子中途消失，猫出现得太早，血腥伤人，多余手指，字幕提前，水印，真实品牌logo，画面糊乱，未成年人形象。

Cinematic high-octane action sequence in a vast red-rock desert canyon under bright blue sky. A heavily customized bright magenta/pink off-road truck with spiked armor, heavy chains, and a massive metal skull front grille with glowing headlights and long fangs speeds through dusty terrain. Dynamic camera work: low-angle undercarriage shot as the truck launches over a dirt ramp kicking up dust and debris; side tracking shot of the spiked truck racing; close-up inside the cab of a stern woman in a traditional yellow hanbok-style outfit smoking a cigarette while gripping the wheel; rear gunner platform with a young woman in digital camouflage military uniform (name patch visible), long hair flying, sunglasses on head, enthusiastically firing a large mounted heavy machine gun with muzzle flashes while shouting; POV from the truck’s spiked hood racing toward a charging herd of massive armored metallic rhinoceros-like beasts with mechanical plating and horns kicking up dust; side and frontal shots of the pink skull-truck barreling through the stampeding armored rhinos as the gunner keeps firing; high aerial overhead shot of the truck weaving among the herd leaving dust trails; final rear tracking shot of the truck speeding away down the canyon with the gun still elevated. Fast-paced, dusty, intense Mad Max-inspired post-apocalyptic energy mixed with Korean action-comedy flair, bright desert lighting, motion blur, practical dust and debris effects, 15-second continuous action sequence.

15 秒，竖屏 9:16，iPhone 原生相机质感，单个连续镜头，一镜到底，无剪辑、无跳切、无转场。参考视频只用于参考“前半段正面悬浮表演、后半段向右绕行揭露机关”的节奏和运镜，不复制参考视频中的场地、落地支架、顶部字幕、水印和人物。 场景是白天的城市室外人行道，天气明亮但不过曝，柔和自然日光，真实手机随手拍质感。两位成年女性像关系很好的闺蜜一样，在路边临时开一个轻松的小玩笑。人行道平整、干净、空间开阔；主要背景是一面无反光的浅色建筑实墙和少量自然绿化，不出现临街玻璃橱窗、镜面门、反光金属墙。远景偶尔有一两名普通成年路人自然横向走过，路人距离主体至少 4 米，不停留、不围观、不看镜头、不模仿动作，也不穿过咖啡杯、悬臂、@1 与 @2 所在的核心表演区域。画面核心道具始终只有一个咖啡杯、一根灰色水平悬臂和一张白色纸条。 【从第一帧已经搭建好的唯一实物机关】 Exactly one rigid straight gray horizontal boom arm, parallel to the floor, extends forward from @2’s waist toward the camera, with the coffee cup fixed at its front tip. The same boom already exists at frame 1 and never appears, grows, extends, changes direction, or changes shape later. 唯一的一根灰色悬臂长约 80 厘米、直径约 2.5 厘米，始终笔直并与地面平行，保持在 @2 腰部至胸腹部高度。悬臂前端只连接咖啡杯背面正中央，连接点完全藏在杯后，杯子左右两侧没有突出的短杆或接头；后端最后约 10 厘米被 @2 紧贴腰腹的前臂和衣袖完全遮住并固定。悬臂不接触地面、墙壁或天花板。杯底到地面之间始终是清楚可见的完整空地。 开场时，镜头光轴、咖啡杯中心、灰色悬臂长轴、@2 躯干中心严格处在同一条景深直线上。咖啡杯从正面完整遮住其后方的悬臂，因此机关虽然从第一帧就真实存在，正面却完全看不见；后续只因摄影机向右绕行产生视差，才逐渐看到同一根悬臂，绝不是中途生成一根新杆。 中号蓝白外带咖啡杯，皇家蓝杯身、白色杯盖，杯身正中只有清晰白色英文“JOHN”。没有真实品牌 and 鹿角图案。杯子全程固定在同一个空间位置，保持竖直，大小不变，不漂移、不旋转、不变形。全片只有这一个杯子。 白色纸条宽 14 厘米、高 10 厘米，从第一帧起就已经用两小段透明胶带固定在悬臂中段的下侧。纸条顶边贴杆，纸面所在的竖直平面与悬臂长轴平行：开场正面看时纸条严格侧对镜头，只露出极窄纸边；镜头向右绕到侧面后，纸条正面才随视差自然展开。纸条只写两行，第一行不得拆行： 您的广告位 hf_20260723_023253_17d01f18-b189-45de-be9c-f30db6d27116 @john87445528 @1 穿 @1 参考服装 图片_20260710063944_1206_183(1) ，始终站在画面左侧三分之一处，双脚原地不动，身体 and 头部绝不越过画面中线，绝不走到咖啡杯正前方或咖啡杯与 @2 之间。@1 只从画面左侧把双手伸向杯子上下方表演；手掌与杯子始终保持约 10 厘米距离，不接触杯子。她像在街边逗闺蜜开心一样俏皮、活泼、聪明灵动：自然放松地轻微侧身，柔软翻腕，十指缓慢张开。她的眼神在杯子、@2 与镜头之间自然移动，偶尔轻挑一侧眉毛，露出忍不住想笑的顽皮笑容，并在成功展示“悬浮”时给镜头一个短促俏皮的眨眼。不是正式商业摆拍，不端着，不低俗，不夸张扭动，不做鬼脸。 @2 穿 @2 参考服装 图片_20260710063950_1209_183 ，双脚不动，始终站在咖啡杯正后方的同一景深轴线上。@2 用贴近腰腹的前臂和衣袖隐藏悬臂后端，固定机关的手臂、身体 and 站位全程不动；但在脸部仍清楚可见的前 5 秒里，她像正在配合闺蜜恶作剧：先故作认真地盯着杯子，随后眉毛上扬、眼睛略微睁大，明知机关却装作吃惊；看到 @1 吹气时短暂抿嘴憋笑，再以“差点被你骗到”的无辜幽默眼神看一眼 @1 和镜头。两人之间有熟悉、放松的默契感。@2 只做面部和轻微眼神变化，不说话、不张大嘴、不摇头、不耸肩。5 秒后镜头推近，@1 与 @2 的脸自然被裁出画面。 【严格连续时间轴】 0—2 秒：正面中全景，摄影机完全不横移。杯子位于画面中心，@1 固定在左侧，@2 位于杯子正后方。杯底下方完整人行道路面清楚可见。@1 一手停在杯口上方、一手停在杯底下方，不接触；她先故作认真地看杯子，再用眼角看向 @2 和镜头，轻挑一侧眉毛并露出忍笑的顽皮表情。@2 配合闺蜜先认真盯着杯子，随后眉毛慢慢抬起、眼睛略微睁大，装出“居然真的浮起来了”的幽默反应。两人的身体 and 脚都不移动。远处可以有一名路人自然经过，但不看她们，也不进入主体轴线。 2—5 秒：摄影机仍在同一正面位置。@1 的脚 and 身体不移动，只做三组清楚、缓慢的手部动作：右手从杯口上方横向掠过；左手从杯底下方横向掠过；双手向外张开展示空手。她像在故意逗 @2，嘴角逐渐忍不住上扬；随后从左侧靠近杯口轻轻吹气，吹气后先看 @2 一眼，再看向镜头短促俏皮地眨一次眼，最后带着“怎么样，厉害吧”的得意小笑容用双手一上一下框住杯子。@2 的身体 and 固定悬臂的手臂保持绝对不动，只用表情接梗：先在杯子 and @1 之间来回看一眼，@1 吹气时抿嘴憋笑，最后用“我就静静看你演”的无辜幽默眼神看向 @1。两人表现得像熟悉的闺蜜自然打趣，不像演员背台词。整个 3 秒内杯子、@2 与摄影机光轴保持共线，悬臂完全被杯子遮住。 5—7 秒：摄影机只沿光轴缓慢向前推近，不横移、不旋转，从中全景推进到距离杯子约 45 厘米的近景。自动对焦锁定“JOHN”。@1 and @2 的脸逐渐离开画面，只保留杯子、@1 的双手、少量服装边缘以及杯底下方的空地。悬臂仍因共线遮挡而不可见。 7—10 秒：摄影机以咖啡杯为固定旋转中心，在同一高度、同一距离缓慢均匀向右绕行约 70 度，用完整 3 秒完成，不甩镜。杯子 and 悬臂都不动。7—8 秒只因轻微视差露出杯后的一小截灰色横杆；8—9 秒看到约一半；9—10 秒看到同一根完整悬臂从杯背面笔直延伸至 @2 腰腹方向。它在画面中呈水平横线，纸条也从极窄纸边逐步展开。人物面部不再入镜。 10—12 秒：保持同一个镜头，摄影机在悬臂侧面沿着悬臂从杯子前端缓慢移动到中段。咖啡杯自然退到画面左边缘，唯一的灰色横杆贯穿画面中央，纸条正面逐渐转到镜头前。焦点从横杆平滑转移到纸条，纸条不是突然出现。 12—15 秒：摄影机到达纸条近景后停止位移并稳住。纸条约占画面宽度 50%，灰色悬臂从纸条背后水平贯穿，咖啡杯只在左边缘保留少量。两行文字稳定、端正、清晰可读：“您的广告位 hf_20260723_023253_17d01f18-b189-45de-be9c-f30db6d27116 .”

CAMERA: Modern smartphone-style handheld selfie POV, first-person vlog. SEOYUN always holds the camera. Natural hand shake, imperfect framing, quick reframes, occasional face cropping. The camera device is never visible. LOOK: Clean modern digital video, sharp and vibrant, natural skin tones, bright even exposure. Full-frame image filling the entire screen edge to edge, standard flat rectangular frame. No black borders, no rounded corners, no letterboxing, no vignette, no lens distortion, no fisheye, no frame overlay, no retro or tape effect. STYLE: Bright, playful K-pop idol vlog. Genuine smiles, eye contact, candid energy, quick handheld shots with backstage excitement. CHARACTER: SEOYUN, Korean female idol in her early 20s. Long straight jet-black hair with soft face-framing strands, dewy fair skin, large dark brown eyes, natural coral-pink lips, gentle sweet smile. Slim figure. Wearing a futuristic white K-pop stage outfit: an asymmetric white halter-neck crop top with one iridescent holographic sequin panel and a thin silver sparkly shoulder strap, a white pleated mini skirt with holographic sequin pleats and decorative silver zippers, a wide white belt with a large rhinestone buckle, long white gloves, white thigh-high boots, a delicate rhinestone choker necklace, and sparkly drop earrings. Stage glam makeup. SETTING: Backstage hallway of a music show studio with dressing room doors. One dressing room door has a clean printed name plate reading exactly "SEOYUN" in bold uppercase Latin letters. Snack table with an open bowl of small individual snacks, overhead PA speaker, and blurred staff passing by. SEQUENCE: 1. Selfie: peeks into the hallway. "얘들아, 우리 지금 무대 뒤에 있어!" 2. Whip pan down the corridor. "이 복도 진짜 길다!" 3. Stops at her dressing room, points at the name plate that clearly reads "SEOYUN". "이거 나야!" 4. Close-up of the name plate, the text "SEOYUN" sharp and clearly legible in bold uppercase letters as she taps it. Ambient audio only. 5. Quick outfit check with a spin near a mirror. "나 오늘 어때?" 6. At the snack table, picks up one small snack piece from the bowl with her fingers and pops it into her mouth. "이건 못 참지." 7. Mid-bite, PA announcement interrupts. She laughs: "잠깐만—" 8. Selfie ending: hears the countdown, waves, makes a hand heart, then runs toward the stage. "이제 무대 가야 돼—무대 끝나고 보자!"

Genre: Cinematic comedy, deadpan humor. Duration: 8 seconds. Tone: Begins completely realistic and calm, then escalates into absurd comedy with perfect timing. IMAX cinematic quality, ultra realistic, Hollywood blocking. ## Scene A modern apartment bedroom at night. Soft blue moonlight enters through the window while a warm bedside lamp creates cinematic contrast. White bedsheets, gray pillows, minimalist bedroom, realistic shadows, peaceful atmosphere. ## Subject ### Image Reference 1 A beautiful young woman, approximately 24 years old, with an expressive oval face, fair skin, soft natural makeup, long dark brown wavy hair, warm hazel eyes, and delicate facial features. She wears a loose oversized black T-shirt and dark gray lounge shorts, barefoot, realistic skin texture with visible pores, slightly messy hair from sleeping, subtle sleepy eyes, documentary realism, ultra photorealistic 8K. ### Image Reference 2 AAA, a genuine extraterrestrial gray alien. Height approximately 135 cm. Extremely large smooth bald head with realistic proportions. Pale gray skin with subtle green undertones, thin athletic body, long arms and fingers, oversized glossy black almond-shaped eyes, tiny nostrils, small mouth, no ears, highly detailed skin pores and wrinkles, hyper realistic biological textures, friendly facial expression, no clothing, no accessories, Hollywood-quality alien design inspired by classic extraterrestrial sightings, absolutely not cartoon, not cute, not chibi, not stylized. ## One Continuous Sequence ### Moment 1 | 0.0-1.8s Wide IMAX cinematic shot from inside the bedroom, facing the bed from the footboard. The woman sleeps peacefully on the left side of the bed. The room is completely silent. The camera performs a slow cinematic dolly-in. ### Moment 2 | 1.8-2.8s Cut to an extreme close-up of the woman's face. Her eyes suddenly snap open. She stays perfectly still with a blank expression, only blinking twice. Slow cinematic push-in increases tension. ### Moment 3 | 2.8-4.5s POV shot from the woman's perspective. The camera slowly pans right. AAA the gray alien is lying under the blanket beside her, facing her with a cheerful smile while repeatedly waving using one long hand. The alien blinks naturally and silently continues waving as if greeting an old friend. ### Moment 4 | 4.5-5.2s Cut back to a medium shot. Without changing her completely serious expression, a heavy black cast-iron frying pan instantly materializes in her right hand from nowhere. No visual effects. No explanation. Perfect deadpan comedy timing. ### Moment 5 | 5.2-6.8s Ultra-fast IMAX whip movement. She swings the frying pan with maximum force directly into the alien's forehead. Loud metallic impact. The alien is launched dramatically across the room in exaggerated slow motion, spinning several times through the air while still continuously waving with a cheerful smile.

A photorealistic smartphone selfie vlog that looks exactly like a real mobile phone recording. A young woman (image = her face and hair) is exploring a lively shopping street in Japan on a bright afternoon, filming herself in selfie mode with natural handheld movement, subtle hand shake, realistic autofocus, authentic smartphone stabilization, and true-to-life colors. She smiles at the camera and says, "Let's see what we find today!" As she walks through the busy street, she suddenly hears live music and turns the camera to reveal a talented Japanese street musician playing an acoustic guitar while a small crowd gathers around. Curious, she walks closer, stands quietly for a few moments, gently sways to the music, and smiles with genuine enjoyment. When the performance ends, she claps enthusiastically along with everyone else, gives the performer a thumbs-up, then turns the camera back to herself and says, "That was amazing!" She continues walking through the street, gives a warm wave to the camera, smiles brightly, and says, "See you in my next vlog. Bye!" before reaching toward the phone to stop the recording. The video should feel exactly like a genuine smartphone vlog with natural human movement, consistent facial features, authentic Japanese street atmosphere, realistic crowd behavior, no beauty filters, no CGI, no AI-plastic appearance, no subtitles, no logos, no watermark, and no background music—only real street ambience, conversations, footsteps, applause, and the live acoustic performance as part of the environment.

A realistic handheld travel vlog filmed by a friend following the main character throughout the day. Use the woman from the reference image as the main subject. Maintain her exact facial identity, hairstyle, facial features, and body proportions throughout the entire video. The camera feels like a real personal vlog camera, not a commercial production. Natural handheld movement, casual framing, imperfect human camera motion, authentic everyday atmosphere. No scripted acting. The woman behaves naturally, interacting with the environment like a real travel vlog. 0-5s: Morning departure. The woman leaves a cozy apartment with a small backpack. She checks her phone, smiles at the camera, adjusts her hair, and starts walking outside. The camera follows her from behind, slightly shaky like a friend filming. Morning sunlight, quiet neighborhood streets, people starting their day. 5-12s: Exploring the city. The camera follows her walking through local streets. She visits a small cafe, buys a drink, briefly talks to the camera, laughs naturally. She walks through a street market, looks at small shops, takes casual photos. The camera stays close, capturing spontaneous moments. 12-20s: Arriving at the beach. She takes public transportation or walks toward the coast. The environment gradually changes from city streets to a seaside town. Ocean breeze moves her hair. She looks excited when she sees the ocean. The camera follows her walking along the beach. She picks up a seashell, watches waves, and interacts naturally with people nearby. 20-27s: Summer beach afternoon. She meets friends at the beach. Everyone chats, laughs, plays near the water. The camera moves naturally between people, capturing real candid moments. She looks back at the camera and smiles. 27-30s: Ending moment. Golden hour sunset. She sits near the ocean, holding a drink, watching the sunset. The camera slowly moves backward, revealing the beach, waves, and the peaceful evening. A feeling of a real personal travel memory. Visual style: Authentic travel vlog footage. Realistic smartphone or mirrorless camera look. Natural daylight. Casual handheld movement. Slight camera shake. Real human reactions. Documentary realism. No cinematic commercial look. No dramatic posing. No artificial transitions. No text overlays. No logos. No face changes. No identity changes.

Super casual real smartphone home video footage, casual office celebration in a decorated break room, natural mobile phone camera with slight authentic handheld shake, normal frame rate with smooth normal motion, rapidfire montage with quick jump cuts every 1-2 seconds like scrolling through phone memories, unpolished authentic phone recording of colleagues cutting cake, clapping and chatting, pure raw home video feel, no cinematic polish. Use the provided reference photo as the strict ONLY visual reference for the main woman. Maintain her exact appearance with zero deviation. Generate a mixed group of colleagues of all ages around her in the office break room, balloons and a small cake visible. 0-2.5s: Shaky rapid cuts — main woman laughing near the decorated table, balloons in the background, quick flashes of colleagues clapping. 2.5-5s: Abrupt jump cuts — close-up of her smiling while cutting the cake, then colleagues passing cake slices around. 5-7.5s: Fast shaky — she chats animatedly with coworkers, casual office attire, laughter visible. 7.5-10s: Quick cut close-up — warm smile toward camera, then jump to her laughing with a senior colleague nearby. 10-12.5s: Abrupt edit — group gathered around the table, casual toast with paper cups, mixed ages chatting. 12.5-15s: Final rapid transition — main woman relaxed among colleagues, soft smile, calm office memory ending with slight natural phone sway. Natural smartphone video quality, slight real handheld shake, smooth normal frame rate motion, authentic casual physics, stable main character consistency, no pro stabilization or effects.

Use the uploaded reference image as the exact character reference. Preserve her facial identity, hairstyle, eye color, makeup, skin tone, body proportions, outfit, and accessories consistently throughout every shot. Maintain perfect facial consistency, realistic hand anatomy, and natural expressions. Use the same iced coffee bottle with identical branding, label, colors, and packaging throughout the commercial. Create an ultra-realistic premium UGC iced coffee commercial with a warm lifestyle aesthetic. Set the film in a modern apartment filled with soft natural window light, warm wooden interiors, cozy neutral décor, realistic reflections, shallow depth of field, cinematic color grading, and premium commercial lighting. The commercial opens with a close-up of the woman smiling at the camera while holding the iced coffee bottle beside her face. She naturally presents the product with an excited expression. Transition to a clean macro product shot where the bottle stands on a wooden table covered with natural condensation, highlighting the premium label and chilled texture. Move into an extreme macro shot as she slowly twists open the bottle cap. Capture the satisfying opening motion and the fresh condensation around the bottle. Show a cinematic macro sequence of the iced coffee being poured over a glass filled with crystal-clear ice cubes. Rich coffee flows smoothly while ice gently clinks together. Capture a close-up of the coffee swirling beautifully as milk blends naturally, creating smooth marbled patterns inside the glass. She lifts the drink, takes a refreshing sip, closes her eyes briefly, and smiles naturally as she enjoys the taste. She then looks directly into the camera while holding the bottle and gives a confident thumbs-up with a satisfied smile. Transition to a clean lifestyle shot where the bottle rests beside a laptop, sunglasses, and a book on a wooden table, creating a relaxed work-and-coffee atmosphere. The next scene follows her walking outdoors under warm sunlight while holding the iced coffee bottle naturally, enjoying the fresh air with a relaxed smile. Finish with a premium hero product shot. The iced coffee bottle stands on a wooden surface surrounded by roasted coffee beans as warm sunlight illuminates the bottle. The woman gently places her hand beside the product, smiles toward the camera, and says, "Good coffee. Good mood. Anytime." The camera slowly pushes in on the bottle before fading out. Style: Premium UGC commercial, luxury café aesthetic, cinematic lifestyle storytelling, realistic coffee textures, macro product cinematography, smooth handheld and gimbal camera movements, commercial-grade lighting, shallow depth of field, photorealistic, ultra-detailed. Negative Prompt: No text, no subtitles, no logos beyond the product packaging, no watermarks, no duplicate people, no distorted hands, no unrealistic facial features, no AI artifacts, no cartoon style, no flickering, no oversaturated colours, no low-resolution textures.

【风格】巨人之手娃娃穿搭（GWRM · Giant Hands Barbie Doll Styling），真实手机竖屏实拍质感（9:16 Vertical, Photorealistic），ins爆款粉色少女感，自然家居光\n【时长】15秒\n【场景】真实居家客厅：浅色木地板、白色衣柜、墙上挂着电视，自然窗光从侧面打进来；女主以真人芭比娃娃的比例站在地板上，画外伸进来的手比她整个人还大\n【角色】女主@图片1（被缩小成娃娃比例的真人，面部与身材保持真人质感，动作有轻微娃娃感的僵直和顿挫）\n【核心机制】比例差：一双巨大的手（正常人手，相对女主如巨人）从画外反复伸入画面为她换装——捏起、放下、脱衣、套衣、蹲下换鞋；女主全程配合但保持娃娃式的乖巧站姿，被摆弄时身体轻微晃动\n【常驻字幕】画面上方粉色手写花体英文标题\"GWRM\"，下方小字\"Styled by Giant Hands Like a Barbie Doll\"，配粉色蝴蝶结与星星装饰\n\n[00:00-00:02] 开场：把娃娃放下（Placement）\n巨手捏住女主的腰部把她从画外拎进来，轻轻放到木地板上站稳，她穿着粉格睡衣外套、白色吊带、白色蕾丝短裤和白袜，头上戴粉色睡眠眼罩，双脚落地时轻轻晃了一下才站定。\n\n[00:02-00:05] 第一步：摘眼罩、脱外套（Undress）\n巨手两指捏住她头上的粉色眼罩向上一摘；随后双手捏住粉格外套的两只袖口，把整件外套从她身上向后一褪、整件拎出画面，她只剩白吊带和蕾丝短裤，抬手揉了揉眼睛。\n\n[00:05-00:08] 第二步：护肤上妆（Skincare Close-up）\n切怼脸特写：巨大的手指蘸着乳白色面霜在她脸颊上轻轻抹开，她乖乖闭眼仰着脸；接着一根手指举起一面粉色小镜子递到她面前，她睁眼照镜子、抿嘴整理刘海。\n\n[00:08-00:11] 第三步：换新装（Dress Up）\n切回全景：巨手捧着一件绿色荷叶边抹胸上衣从她头顶套下、拉正下摆；再取一条粉白格子超短裙从她脚下向上提到腰间，抚平褶皱，她配合地抬起手臂又放下。\n\n[00:11-00:13.5] 第四步：换鞋袜（Shoes & Socks）\n低角度贴地特写：巨手捏着粉色长筒蕾丝袜一只只往她腿上套、拉到膝盖；再把一双玫红色尖头高跟鞋摆到她脚边，捏着她的脚踝一只只塞进鞋里，指尖调整鞋跟站稳。\n\n[00:13.5-00:15] 收尾：走秀定格（Final Pose）\n巨手退出画面，女主独自站在地板中央，双手叉腰扭胯摆pose，接着原地转身甩了下头发，对镜头歪头一笑，画面定格在她的全身穿搭上。\n\n【音效】轻快甜美的ins风BGM打底，衣物摩擦的窸窣声、鞋跟磕地的\"哒\"声、手指擦面霜的细微声都保留真实录音质感，每次换装完成配一声轻\"叮\"，结尾定格配上扬收束音