AI Prompt Library

Discover and copy the best AI prompts for image, video, and audio generation — tested on GPT Image 2, Nano Banana, Seedream 5, Seedance, MiniMax H3 and more.

Create a 17-second vertical 9:16 video at 24fps in cinematic 4K photorealism. Reproduce the exact comedic sequence, actions, timing, character positions, and countertop layout from Video 1. Use one continuous, locked-off shot with absolutely no camera movement, zoom, or cuts. CRITICAL OBJECT CONTINUITY: There must be EXACTLY ONE cup throughout the entire video. It must remain the same physical object with identical appearance, shape, and size. The cup can only move through visible, continuous hand contact. Show every pickup, movement, placement, and release. Never duplicate, teleport, slide, or make it disappear. The coffee jar and spoon must also remain single, persistent objects. ACTION SEQUENCE: 0–5 SECONDS | FIRST MIX-UP Character A (Image 1) looks exclusively at the coffee jar and spoon while scooping dry coffee powder. At the same time, Character B (Image 2) picks up the ONLY cup, carries it to their side, pours hot water, and places it down. Unaware of this movement, A pours coffee powder toward the cup's original position, spilling it onto the countertop. 5–10 SECONDS | SECOND MIX-UP A continues looking only at the coffee jar and spoon. B picks up the SAME cup, physically carries it back to A's original position, and places it down. A mistakenly pours coffee powder toward the cup's previous position, spilling it onto the countertop again. 10–17 SECONDS | FINAL PUNCHLINE The countertop must be completely clean and dry in the area where milk will land. A physically picks up the SAME cup. At that exact moment, B starts tilting the milk container toward the cup's original position. Milk must visibly exit the container as a continuous stream. Since A has lifted the cup, the milk lands directly on the countertop and progressively forms a natural spill. The instant A notices the milk spilling, A reacts with unmistakable shock: a sudden upper-body jerk, brief wobble, wide-eyed astonishment, and a genuinely funny, disbelieving expression. B looks at the spilled milk with a helpless, awkward, mildly frustrated expression. STRICT GAZE CONTROL: During both coffee-scooping sequences, A must continuously keep their head lowered toward the coffee jar. A's eyes must remain fixed on the jar and spoon. The head and gaze must NEVER follow the spoon toward the cup or look at B. Only A's arm and hand may move toward the expected cup position. CHARACTER BEHAVIOR: Both characters remain unaware of each other's actions throughout the three accidental mix-ups. All hand movements and object interactions must be anatomically correct, physically realistic, and continuous. COFFEE POWDER CONSISTENCY: The jar contains ONLY dry, granular coffee powder throughout the video. Every scoop must visibly collect dry powder. Powder must adhere to the spoon and fall naturally. Spilled powder must remain completely dry. Never transform it into liquid, slurry, or another substance. MILK PHYSICS: Milk must originate ONLY from B's container. Maintain this exact order: Dry countertop → container tilts → visible milk stream → milk lands → spill accumulates. Never generate milk spontaneously or before pouring begins. FINAL REQUIREMENTS: No extra cups, spoons, ingredients, gestures, or events. Prioritize exact object identity, physical continuity, realistic hand interactions, character consistency, and perfectly timed comedic reactions over visual embellishments.
Video 1: Use exclusively for motion, actions, facial expressions, hand gestures, selfie framing, timing, and camera movement. Image 1: Use exclusively for the woman's identity, face, makeup, hairstyle, outfit, and environment. Never allow the video reference to override the identity or appearance of the woman in Image 1. CHARACTER IDENTITY LOCK: Replace the original woman in Video 1 with the woman from Image 1. Preserve her exact facial identity in every frame, including face shape, eyes, eyebrows, nose, lips, jawline, skin tone, and makeup. No beautification, facial alterations, aging, face blending, or identity drift. HAIRSTYLE AND OUTFIT: Maintain the exact hairstyle from Image 1: - Long, voluminous, layered blonde wavy hair. - Hair length ending just below the chest. - Never extend below the waist. Preserve the exact outfit: - White low-cut cropped short-sleeve knit top. - Light-blue washed low-rise denim mini skirt. MOTION REPLICATION: Follow Video 1 precisely throughout the entire video. Reproduce all original: - Hand gestures and movements. - Facial expressions and blinking. - Hair-flipping movements. - Blowing kisses and face-touching gestures. - Timing and movement intensity. - Selfie framing and camera trajectory. Do not invent or remove actions. HAND ANATOMY: Both hands must remain anatomically correct during every gesture. All five fingers must appear natural and properly formed. No extra, missing, fused, or distorted fingers. ENVIRONMENT: Replace the original video environment entirely with the exact background from Image 1. Maintain consistent soft lighting and warm ambient illumination. No additional people anywhere in the scene. VIDEO QUALITY: - Clean, full-screen video. - No logos or watermarks. - No text or subtitles. - No UI overlays. - Natural movements and realistic motion blur. FINAL REQUIREMENT: From the first frame to the last, maintain perfect facial identity consistency with Image 1. Image 1 controls identity, appearance, outfit, and environment. Video 1 controls motion, expressions, gestures, timing, and camera movement. Never mix or exchange their responsibilities.
Create an ultra-realistic cinematic sci-fi fantasy action sequence featuring a beautiful white-haired female warrior with a black eye mask, wearing a sleek futuristic black armored bodysuit and a flowing black cape. Scene 1: She jumps from a futuristic aircraft into the open sky, holding dual futuristic energy pistols. Her long platinum blonde hair flows dramatically in the wind. Scene 2: She flies through a massive dark storm dimension filled with enormous purple lightning clouds, supernatural energy and floating debris. Her armor activates with glowing neon green energy patterns. Scene 3: Multiple terrifying black winged creatures emerge from the storm. She engages in an intense aerial battle, firing powerful green energy blasts from her dual pistols. Scene 4: Show dynamic close-up shots, first-person weapon perspectives, high-speed aerial movements, dramatic camera rotations and cinematic slow motion as she dodges enemy attacks. Scene 5: She summons a glowing green magical circle beneath her feet while fighting a gigantic dark armored shadow creature with glowing purple eyes. Final scene: A massive purple energy portal opens in the sky. She flies through it and lands gracefully on top of a futuristic black aircraft. Visual style: Hollywood blockbuster CGI, photorealistic 3D rendering, Unreal Engine 5 quality, highly detailed character design, realistic hair and fabric physics, volumetric clouds, dramatic cinematic lighting, purple and neon green color palette, motion blur, IMAX quality, 4K, 16:9 widescreen. Maintain consistent character appearance, armor, hairstyle and facial features throughout the entire sequence. Smooth transitions, fluid animation, epic action choreography.
Create an ultra-photorealistic, cinematic 30-second Japanese countryside lifestyle video featuring a young Japanese woman enjoying a peaceful morning in a traditional rural village. The video opens with a beautiful young woman stepping out of a traditional Japanese wooden farmhouse. She wears a simple cream-colored blouse, a flowing olive-green skirt, and carries a small woven shoulder bag. Capture her gentle smile as she opens the wooden sliding doors and steps into the warm morning sunlight. Next, show her walking along a quiet village pathway surrounded by traditional wooden houses, stone walls, potted plants, and lush green mountains. She meets two elderly village women tending their beautiful flowerpots and shares a warm, friendly conversation with them. Transition to a charming countryside bakery with rustic wooden interiors. The young woman joins a lovely elderly baker behind a wooden counter filled with freshly baked golden croissants and pastries. Capture a heartwarming moment as they smile and share the freshly baked bread. Next, show her standing near a traditional Japanese railway crossing. A beautiful vintage green-and-cream local train slowly passes through the scenic countryside. Capture the train, railway barriers, lush mountains, and peaceful village atmosphere from behind the woman. The final sequence shows her sitting at a charming outdoor wooden café surrounded by blooming pink flowers. She holds a delicate ceramic cup of warm tea, takes a small sip, and enjoys the peaceful morning sunshine. Visual Style & Quality: * Ultra-photorealistic live-action cinematography * Authentic Japanese countryside setting * Natural facial expressions and human movements * Warm, soft morning sunlight * Traditional Japanese architecture * Lush green mountain landscapes * Vintage green local train * Cinematic close-ups and wide establishing shots * Shallow depth of field * Natural skin texture and realistic details * Soft atmospheric film grain * 4K HDR cinematic quality * Vertical 9:16 aspect ratio Overall Mood: A heartwarming, nostalgic Japanese countryside morning filled with simple pleasures, friendly village encounters, fresh bakery moments, and peaceful everyday beauty.
Create a 30-second high-quality Japanese 2D anime fantasy short film in 16:9 widescreen, using the uploaded image as the exact visual and character reference. TITLE: THE SECRET STRAWBERRY GARDEN CHARACTER CONSISTENCY: The pink-haired girl in the reference image is the main heroine. Preserve her exact face, blue eyes, short pink bob hairstyle, cream-colored round ear accessories, white camisole and blue-white striped shorts. Keep her appearance identical in every scene. Preserve the original anime art style and beautiful kitchen environment. VISUAL QUALITY: Premium Japanese anime movie quality, vibrant strawberry reds, beautiful pinks, rich midnight blues, warm golden kitchen lighting, clean cyan outlines, excellent color contrast, detailed backgrounds and smooth natural animation. Keep the image colorful but not oversaturated. STORY AND ACTION: 0–3 SECONDS: The girl stands in the same kitchen, opens the refrigerator and discovers an unusually large glowing red strawberry. She picks it up with both hands and looks surprised. 3–6 SECONDS: She places the giant strawberry on the wooden countertop and carefully cuts it in half with a kitchen knife. A beautiful miniature garden is revealed inside, filled with tiny green trees, colorful flowers and miniature strawberries. Her eyes widen in amazement. 6–10 SECONDS: She brings a transparent round glass dessert mold to the counter. She pours beautiful translucent pink jelly into it, creating smooth swirling patterns. 10–14 SECONDS: She carefully places the miniature strawberry garden centerpiece inside the transparent dessert. She gently adds fresh cream and small strawberry pieces around it. 14–19 SECONDS: She adds another layer of pink transparent jelly. The dessert gradually becomes a beautiful clear strawberry-shaped dome, with the miniature garden visible inside. 19–23 SECONDS: She decorates the top with fresh strawberries, tiny white flowers and delicate green leaves. The finished dessert sparkles under the warm kitchen lights. 23–27 SECONDS: She carefully cuts a slice from the transparent dessert. The inside reveals an astonishing miniature garden, with tiny trees, flowers and red strawberries surrounded by glowing pink jelly. 27–30 SECONDS: She tastes a small piece, closes her eyes and smiles happily. The camera slowly pulls back to show the beautiful magical dessert on the countertop, with the nighttime city glowing through the kitchen window. BACKGROUND MUSIC: Original gentle Japanese fantasy anime soundtrack throughout. Begin with mysterious soft piano, introduce sparkling magical bells when the miniature garden appears, use playful orchestral music during dessert preparation and finish with a beautiful emotional musical flourish. Include subtle synchronized sounds of the refrigerator, knife cutting, jelly pouring, utensils and gentle kitchen ambience. No dialogue, no vocals. STRICT RULES: - Use the uploaded image as the primary character and visual reference. - Keep the same heroine, face, hairstyle, outfit and accessories throughout. - Maintain the same kitchen layout and nighttime atmosphere. - Keep the strawberry and miniature garden visually consistent. - Use natural hand movements and physically believable actions. - Every action must follow the previous action logically. - Smooth camera transitions, clear compositions and stable objects. - No character morphing, extra fingers, duplicated objects, flickering or random scene changes. - No photorealism, no 3D CGI, no text, no subtitles and no logos. Prioritize character consistency, beautiful color contrast, smooth storytelling and a spectacular final reveal.
EXACT 30 SECONDS Aspect Ratio: 16:9 Create an ultra-photorealistic live-action cinematic travel vlog featuring the exact male identity from @image1 exploring New York City, USA. Preserve his exact facial identity, hairstyle, beard, skin tone and body proportions. Maintain the same olive-green travel jacket, white T-shirt, black cargo trousers, white sneakers, black backpack and wristwatch in every shot. Use the facial close-up panels of @image1 for identity consistency and the headless full-body panels for wardrobe and body proportions. Create a professional 30-second travel documentary with 10 visually distinct shots, realistic human performance, cinematic multi-angle cinematography and smooth narrative continuity. 00:00–00:03 — Times Square: Medium-wide 28mm handheld backward tracking shot. The vlogger smiles and gestures toward enormous digital billboards while yellow taxis and pedestrians move naturally behind him. 00:03–00:06 — Brooklyn Bridge: Wide 24mm stabilized backward tracking shot as he walks along the pedestrian walkway with historic stone towers, suspension cables and Manhattan skyline. 00:06–00:09 — Statue of Liberty: 50mm cinematic push-in at a New York Harbor viewpoint. The vlogger points toward the monument while realistic ferries and water reflections appear in the background. 00:09–00:12 — Manhattan Streets: Dynamic 35mm lateral tracking shot. A yellow taxi passes naturally as the vlogger walks and speaks to camera. 00:12–00:15 — Central Park: Smooth 35mm Steadicam push-in as he explores tree-lined pathways with green landscapes and distant skyscrapers. 00:15–00:18 — Vlogger Close-Up: Intimate 85mm close-up with sharp eye focus, realistic facial expressions, natural dialogue and soft green background. 00:18–00:21 — Empire State Building: Dramatic 24mm low-angle upward tilt revealing the iconic skyscraper as the vlogger looks upward and gestures toward it. 00:21–00:24 — Midtown Manhattan: 50mm smooth backward tracking shot as he walks through towering skyscrapers while speaking directly to the audience. 00:24–00:27 — Brooklyn Waterfront: Spectacular 35mm semicircular 180-degree orbit revealing the Manhattan skyline across the East River during golden hour. 00:27–00:30 — Final Closing: 40mm slow cinematic dolly-out as the vlogger walks beside the waterfront, turns toward the camera and smiles with the glowing Manhattan skyline behind him. Fade to black exactly at 30 seconds. Cinematography: Professional cinema-camera aesthetics, realistic lens characteristics, natural depth of field, smooth physically achievable camera movement, cinematic composition, authentic motion blur and subtle film grain. Lighting: Natural daylight transitioning subtly into golden-hour warmth. Realistic shadows, physically accurate reflections and consistent environmental lighting. Audio: Original upbeat cinematic travel music combining acoustic guitar, electronic percussion, bass and orchestral strings. Gradually build the score, soften it during dialogue, reach an emotional peak during the skyline reveal and finish with a warm orchestral chord. Include authentic New York traffic, yellow taxi engines, horns, pedestrians, footsteps, harbor breeze and river ambience. Natural synchronized dialogue with clear voice priority. Performance: Natural walking, realistic hand gestures, authentic facial expressions, subtle eye movement, confident travel-vlogger personality and believable interaction with the environment. Continuity: Same face, beard, hairstyle, clothing, backpack, accessories and body proportions in all 10 shots. Maintain realistic geography, consistent lighting and seamless visual progression. No random locations or unnecessary characters.
**IDENTITY:** Use image as the exact face, identity and outfit reference. Exactly ONE So-Yeon throughout. Real human actor, authentic skin, eyes, hair and biomechanics. Never CGI or digital double. **SETTING:** Empty late-night Korean subway platform. Metal benches, yellow safety strip, flickering fluorescent lights, tiled walls, arrival board reading "도착 정보 없음", dark tunnel and approaching train. Five unarmed delinquent boys in untidy school uniforms. **STYLE:** Ultra-photorealistic live action, restrained VFX, intimate cinematic framing. Korean dialogue. No music or subtitles. Only diegetic sound. **POWER — HOLLOW NOTE:** So-Yeon awakens into emotionless, close-range Wing Chun combat: CHAIN PUNCH → TRAP → PALM → ELBOW → PARRY → HOOK → KNEE. Minimal footwork, tight centerline strikes, no kicks, spins or dashes. Every physical impact cancels sound at its contact point: IMPACT → ABSOLUTE SILENCE (0.2s) → AMBIENCE AND DELAYED REACTION RETURN. Larger strikes create larger silence zones. Final strike silences the entire platform for 0.35s. Subtle air shimmer only; no glow, particles, shockwaves, wind, heat or supernatural visual spectacle. Horror is entirely audio-driven. **ACTION TIMELINE:** **00:00–00:06:** So-Yeon sits hunched on a bench as five boys surround, slap her head, pull her ponytail, kick the bench and knock her bag toward the tracks. Leader: "야, 반응 좀 해봐. 재미없게." She whispers: "아… 하지 마…" Train rumble swells, then ALL sound abruptly disappears. **00:06–00:12:** Sound returns. So-Yeon stands perfectly still, eyes cold and unfocused. Leader swings; she traps his wrist and palm-strikes his chest. Contact creates localized silence; his soundless gasp arrives late with the returning ambience. She stares at her hand, confused. A bully shouts: "저 새끼가?!" Four rush her. **00:12–00:22:** Rapid, brutally efficient Wing Chun against simultaneous attackers. Chain punches, traps, elbow, parry, palm, hook and knee. Feet barely move. Every readable impact creates a separate silent pocket; overlapping silences swallow the platform as the boys collapse one after another. No rage, no wasted motion. **00:22–00:27:** Leader escapes toward the stairs, then turns and swings wildly. So-Yeon calmly walks inside his attack and lands one powerful centerline palm strike. Entire platform goes silent for 0.35s. He slides backward across the tiles. **00:27–00:30:** Train roar, fluorescent hum and delayed body impact slam back simultaneously. Headlights illuminate So-Yeon's blank face. She studies her open hands, briefly frightened, then walks toward the arriving train. Never looks back. CUT. **CAMERA:** Tight, steady close-ups, slow push-ins during silence, brief handheld jolts on contact. No whip-pans, speed ramps or chaotic editing. Hold each silence uncomfortably long. **NON-NEGOTIABLE:** One consistent real human. Clearly readable simultaneous combat. Every sound cancellation follows physical contact. No superhero aesthetics. Unsettling realism through impossible silence.
Create a cinematic, ultra-realistic, humorous action-comedy video set on a rooftop overlooking an urban neighborhood, with natural daylight and highly detailed environments. A beautiful young Asian woman with short black hair is lying beside a large, fluffy white Persian cat on a rooftop table. In front of them is a realistic sniper rifle with a telescopic scope. The woman looks excited and playful while the fluffy white cat sits calmly beside the rifle, appearing extremely serious and focused. The woman aims the rifle toward a nearby narrow alley. Suddenly, a strange glowing golden fish appears floating in midair, creating a surreal and unexpected moment. Cut to the alley, where an orange tabby cat is eating fish from a blue food bowl near a metal trash can. A muscular, dark-colored dog suddenly appears in the distance and starts approaching the orange cat. Switch to a cinematic sniper scope POV shot, showing the dog walking toward the orange cat. The crosshairs lock onto the dog. Suddenly, the dog transforms into an extremely muscular, bodybuilder-like creature with exaggerated muscles and a powerful stance, creating a hilarious unexpected twist. Cut back to the rooftop. The young woman and the fluffy white Persian cat look amused and surprised. The woman smiles and waves her hand. In the final sequence, she puts away the rifle and casually starts feeding her fluffy white cat with small pieces of food from a paper bag, while the cat remains calm and proud. Camera Direction: Use cinematic wide shots, close-ups, realistic sniper scope POV, smooth transitions, and dynamic action cuts. Visual Style: Photorealistic, cinematic lighting, highly detailed fur, realistic animal movements, expressive facial reactions, natural rooftop atmosphere, and absurd comedy. Audio: Natural ambient city sounds, subtle action sound effects, and humorous timing. Aspect Ratio: 16:9 horizontal. Duration: 30 seconds. Maintain consistent character appearances, realistic physics, and smooth scene continuity throughout the video.
A single continuous vertical 9:16 shot, 15 seconds, photorealistic, shot like a viral smartphone/drone-style action video in bright desert sunlight. A motocross rider in a white full-face helmet, a neon-yellow riding jersey and pants, white riding boots and gloves, on a blue dirt bike, rides straight up a gigantic, almost vertical red-orange sandstone wall. A huge crowd stands along the rim at the very top, hundreds of people with phones raised, cheering. Clear deep-blue sky, harsh midday sun, strong warm red and orange tones. Action, in sequence: 0–1s: Close behind-the-rider view. He sits on the bike at the base of the cliff, back to camera, revving, bright yellow jersey against the red rock. Another rider on a blue bike waits in the far left background. 1–2.5s: He launches forward and hits the base of the wall. The rear wheel spins hard and throws a thick, billowing cloud of orange dust. The camera pulls back as he goes up the rock. 2.5–5s: He rides straight up the wall, getting smaller as he climbs. A long fiery-orange dust trail streams down behind him along the rock face. The top rim comes into view with a thin line of spectators looking down. 5–10s: The camera tilts up and tracks him from below as he climbs, bike nearly vertical, back to camera, arms slightly out, wheel kicking up bursts of orange dirt that tumble down the rock. The crowd along the top edge grows bigger and clearer as he gets closer. 10–12.5s: The rim fills the top of the frame. Hundreds of spectators lean over the edge with phones held up, shouting and waving. The rider keeps steady throttle, closing the last few meters, dust still streaming below him. 12.5–15s: He reaches the top. The front wheel crests the edge, the crowd surges forward and grabs him and the bike, pulling them over the edge into the crowd, a mass of hands, phones and raised arms around him. Camera: Starts at close shoulder height behind the rider, then pulls back and tilts up with fast, slightly shaky handheld energy, following him up the cliff from below at a steep low angle. Ends at a low-angle shot of the crowd swallowing him at the rim against the blue sky. Wide-angle phone lens, sharp focus on the rider, natural motion blur on the dust. Lighting and look: Strong direct sunlight, saturated orange-red rock, deep blue sky, high contrast, neon-yellow jersey popping against the red wall, glowing backlit dust clouds, crisp texture on the rock face and bike. Audio: A high-revving two-stroke or four-stroke dirt bike engine, tire spin and rocks skittering down, the crowd's roar growing louder toward the top, ending in screams and cheers as he is pulled over. No music. Avoid: cuts, text overlays, logos or readable brand names, cartoon style, distorted bike or body, extra riders in the main action, watermark.
Create a luxurious, cinematic fantasy fashion campaign inspired by an ethereal blue butterfly collection. A beautiful female model with long flowing hair walks gracefully through a futuristic midnight-blue runway, wearing an elegant translucent silver-blue couture gown decorated with delicate floral embroidery, sparkling crystals, and butterfly-inspired details. Cut to an extreme macro close-up of her neckline, revealing a delicate blue butterfly pendant resting against her skin. A breathtaking luminous blue butterfly lands on an intricately designed luxury silver handbag with quilted butterfly-wing textures and sparkling blue gemstones. The scene transitions into a magical enchanted forest at night. The model stands in the center of a circular field of glowing blue butterflies. Thousands of tiny luminous butterflies swirl around her, forming a giant spiraling vortex of sparkling blue light. The butterflies rise dramatically into the sky like a magical waterfall, surrounding her in an extraordinary fantasy spectacle. Show cinematic macro shots of glowing blue butterfly jewelry attached to an ancient tree, followed by an overhead shot of the model surrounded by hundreds of floating luminous butterflies. Capture elegant details of her flowing sheer gown, delicate footsteps, and a close-up of her hand holding the sculptural butterfly handbag. End with an aerial cinematic shot as glowing butterflies fill the enchanted forest and illuminate the night. Visual style: Ultra-realistic luxury fashion commercial, high-end editorial cinematography, magical realism, deep midnight-blue color palette, silver highlights, luminous cyan butterflies, sophisticated couture styling, ARRI Alexa 65, 85mm macro lens, smooth slow motion, dramatic lighting, volumetric light, shallow depth of field, intricate textures, photorealistic 4K, seamless cinematic transitions, premium luxury brand advertising aesthetic.
Create a 15-second ultra-realistic 16:9 landscape video featuring the same 24-year-old Korean woman with long loose dark hair, minimal makeup, an oversized maroon hoodie, and loose casual pants. Maintain consistent facial identity and appearance throughout. Setting: A cozy apartment kitchen on a sunny morning. Authentic early-2000s Sony MiniDV home-video aesthetic, filmed by another person, with slight handheld shake, imperfect framing, natural lighting, and faded colors. 0–3s: She stands at the stove, preparing pancakes in a frying pan. She confidently shakes the pan and prepares to flip a golden pancake. 3–6s: She flips the pancake into the air, but it flies much higher than expected. She looks up with wide eyes, following its unexpected trajectory. 6–9s: The pancake comes down awkwardly and lands folded over in the pan. She stares at it in disbelief, then looks toward the camera with an embarrassed expression. 9–12s: She carefully unfolds the pancake with a spatula, tries to make it look presentable, and adds a few strawberries on top. 12–15s: She takes a small bite, smiles with satisfaction, and proudly gives the camera a playful thumbs-up. Natural, realistic pancake physics, believable cooking movements, consistent character and kitchen. Smooth motion without stuttering, duplicated frames, distorted hands, or sudden transitions. Audio: Gentle frying sounds, spatula scraping, the pancake landing in the pan, and her spontaneous laughter. No music or narration.
Female Character Identity Reference: Image 1 Police Officer Identity Reference: Image 2 Motion and Camera Reference: Video 1 Background Reference: Image 3 Create a 15-second vertical 9:16 video in a realistic handheld smartphone recording style. 1. CHARACTER IDENTITY LOCK Keep the female character 100% identical to Image 1 and the police officer 100% identical to Image 2. Preserve their exact facial features, hairstyles, skin tones, and body characteristics. No identity drift or face swapping. 2. REFERENCE CONTROL - Image 1: Female character identity. - Image 2: Police officer identity. - Video 1: Storyline, actions, expressions, poses, camera angles, movements, editing, and timing. - Image 3: Background and environment. 3. TIMED STORYBOARD 00:00–00:02.5: Officer approaches the woman sitting sideways on a broomstick. Film from inside a car through the side window. 00:02.5–00:05: Woman retrieves an identification card from inside her clothing and physically hands it to the officer. 00:05–00:08: Officer inspects the card and physically returns it to her. 00:08–00:10: Woman stores the card back inside her clothing, turns toward the camera, and gives a confident smile. 00:10–00:12.8: Woman prepares for takeoff, grips the broomstick, and accelerates when the traffic light turns green. 00:12.8–00:15: Woman flies away on the broomstick. Camera pans toward the front-right, following her before ending on the road and intersection. 4. STRICT VISUAL REQUIREMENTS - Maintain perfect character identity consistency. - Reproduce the reference video's camera movements and action sequence. - Show realistic physical interactions and continuous card handovers. - Exactly one identification card throughout. - No clipping, teleportation, duplication, floating objects, or deformed hands. - The broomstick and rider must move together naturally. - No additional scenes or unnecessary camera cuts. - Maintain realistic lighting, expressions, movement, and cinematic continuity. CORE RULE: Character identities from reference images, motion and camera work from reference video, and environment from the background image. Prioritize physical realism and continuity in every frame.
Ultra-realistic cinematic street portrait of a beautiful young woman with short, messy dark brown bob hair, natural freckles across her nose and cheeks, soft facial features, expressive hazel eyes, and a calm, slightly melancholic expression. She is wearing an oversized, chunky knitted beige wool turtleneck sweater with a thick folded collar covering part of her chin. She is gently holding a fluffy, light gray Persian cat in her arms. The cat has a round face, dense soft fur, small ears, long whiskers, and striking bright amber-orange eyes, looking directly into the camera. Environment: A cold, foggy city street at night, with a cinematic urban atmosphere. Blurred pedestrians in the background, soft glowing streetlights, and an out-of-focus orange neon sign behind her. Subtle mist creates depth and atmosphere. Composition: Vertical portrait, medium close-up framing, subject centered, the woman looking directly into the camera while holding the cat close to her chest. The cat is prominently positioned in the lower center of the frame. Photography: Shot on an 85mm portrait lens, f/1.8 aperture, shallow depth of field, extremely detailed skin texture, realistic freckles, individual hair strands, highly detailed cat fur, natural facial proportions, soft diffused lighting, cinematic film grain, muted earthy color grading, warm amber highlights mixed with cool greenish-gray shadows. Photorealistic editorial fashion photography, authentic candid street photography, atmospheric cinematic lighting, subtle vintage film aesthetic, professional photography, ultra-detailed, 8K quality. Aspect ratio: 9:16 portrait.
【Title】Eyes Behind the Wall 【Duration】Complete 30-second short film, no segmentation 【Aspect Ratio】21:9 widescreen 【Style, Cinematography & Atmosphere】 1974 Italian gothic horror and sitcom hybrid film, 35mm Techniscope. Coarse grain, soft vintage anamorphic lenses, scattered dust spots, fine scratches, authentic old hotel set. Rainy night, desaturated gray-brown, dark olive green and cold gray shadows. Cold light from the window outlines characters' profiles, the old dressing table lamp provides soft localized warm light. Foreground action is clear, shadows retain detail, no smoke. The first half builds unease through voyeuristic gaze and mirror anomalies; when the ghost eye presses close to the wall hole, shoot it as sincere horror, no comedic sound effects inserted beforehand. The second half creates comedy through the four monsters' clumsy explanations and the ghost turning to watch them. Direct hard cuts, clear camera positions for each shot. Mirror reflections, both ends of the wall hole, character movement and entrances/exits remain consistent. Sensuality is conveyed through the female lead's cool gaze, bedtime ritual and restrained posture; she is fully clothed throughout, no undressing, nudity or changing process shown. 【Reference Images & Character Lock】 Reference images are for design, not first frame. @img1@673e352f-5aa0-4b5e-8946-20389bb6b6f7 locks the four monsters: Mouth Monster: enormous pale deformed head, multi-mouthed torso, black segmented armor. Pharaoh: twin-horned golden crown, ancient Egyptian golden attire, responsible for the excuses. Mechanical Gentleman: black bowler hat, old-fashioned suit, half-mechanical face and mechanical hand. Lobster Monster: red curved-horn head armor, broad thick red carapace, enormous twin pincers. The four monsters' appearance and physique remain consistent throughout, no swapping, no duplication. @img2@9e3c0182-94f5-4080-91b3-d888eeff4d10 only locks the male and female leads' physique relationship; the male lead is slightly taller and noticeably broader. Ignore background text and incorrect height markings. @img3@a433dbf2-1c82-4b52-902e-401aadca3f66 locks the adult female lead's face, side-parted dark wavy long hair, @6f624200-dfbf-4e5c-b498-8a959da64d17 locks clothing and figure and stockings with heels. Add a pair of slender silver drop earrings, a hairpin securing part of the long hair. After the female lead leaves the mirror, she puts on an opaque dark gray long nightrobe off-screen, still fully wearing the original black dress underneath. @img4@6a86c06e-c97f-4073-b17e-ded857db586e locks the male lead's face, messy black curly hair, short beard, enormous belly, leather hunting outfit, gray-green cloak and old leather boots. The longsword stays in its sheath. This film takes place in two adjacent old hotel guest rooms and a public corridor. All characters are explicitly adults. 【Female Lead — Siren-like Posture】 The female lead sits on the edge of the bed, body slightly angled, shoulders relaxed and sunk, neck elongated, chin slightly raised, legs naturally together and angled, calves offset front and back. Half-lowered eyelids, gaze calm, restrained, with a sense of distance. When removing earrings and hairpin, movements are slow, fingers clear, only one prop handled at a time. When bending forward to untie shoelaces, leans naturally forward from the hips, one hand on the bed edge for balance, camera maintains normal medium or full shot, no under-skirt or body-part shots. After discovering the wall hole, no exaggerated screaming, no actively posing toward the hole; she turns the lamp to illuminate the eye in the hole and signal that she has noticed. When standing and walking, feet bear weight normally, no maintained crossed legs. At the end, standing by the door in the nightrobe, weight on one side, shoulders relaxed, coldly watching the four monsters; no exaggerated flirtatious smiling, no ocean or magical effects added. 【Fixed Space — Must Follow This Layout】 Two adjacent guest rooms share an old plaster partition wall approximately fifteen centimeters thick. Room A is the female lead's room: The bed edge is near the shared wall. The dressing mirror is beside the bed, at an oblique angle to the shared wall. There is an old pipe hole approximately three centimeters in diameter on the wall, hole center about one meter twenty from the floor. When the female lead sits on the bed edge, the four monsters looking through the hole from Room B can only see her profile and upper body in the dressing mirror, cannot directly see the whole room. The mirror position is fixed, does not rotate. The female lead can also see the eyes in the wall hole through the mirror reflection. There is a screen on the inner side of Room A. After the female lead leaves the mirror, she walks behind the screen, puts on the hung opaque nightrobe, then enters the public corridor from Room A's door. No dressing process behind the screen shown. Room B is the four monsters' room: The hole is above an empty wooden table against the wall. The tabletop is low, does not block the hole. On the other side against the wall is a tall old bed, space underneath sufficient for one person to crouch and hide, but cannot accommodate all four monsters simultaneously. Room B's door is near the public corridor, ajar at the opening. Room A and B doors are on the same corridor, about two meters apart. The male lead enters Room B from the corridor. The real female lead exits Room A later, walks along the corridor to Room B's doorway, does not pass through walls, does not teleport. 【Mirror Double & Horror Rules】 At the opening, the mirror reflects the female lead's movements normally. After the real female lead leaves the mirror, the female lead in the mirror does not disappear, still sits in place. The double has the same face, hair and black dress, but does not have the gray nightrobe the female lead subsequently puts on, expression cold and stiff. The double first raises its eyes in the mirror, then walks along the mirrored room toward the wall hole's location. When observed from Room B's hole, the four monsters initially still see it in the mirror; then a pale eye directly presses against the Room A end of the hole, blocking the distant mirror surface. Only at this moment has the ghost entered Room A's physical space from the mirror image. Conveyed through two clear shots: First, in Room A's wide shot, the mirror surface in front of the empty bed edge slightly concaves inward, a pale foot lands on the floor; Second, the ghost walks along the wall to the hole's edge, bends down close to the hole. No need to show the complete mirror-crossing process, but must provide evidence of landing and walking toward the wall hole. The ghost hand can only extend one slender finger from the existing wall hole, cannot let the entire palm pass through the three-centimeter hole. The finger extends along the hole's axis, lightly touches the mechanical eye then retracts. The wall does not melt, does not expand. 【Prop Continuity】 The female lead removes one earring at the opening, the other remains on her ear; the hairpin is subsequently removed, placed in a shallow dish on the dressing table. After untying, high heels are placed side by side beside the bed, the female lead subsequently wears indoor flat slippers to walk to the next room, slippers placed visibly under the bed in advance. The original polka-dot stockings remain intact. The male lead holds a coarse ceramic noodle bowl with both hands, chopsticks laid across the rim. The bowl is not duplicated, does not spill soup, hands are not simultaneously freed to operate other props. The four monsters have no IOUs or additional handheld props, avoiding items appearing without setup. 【Strict 30-Second Storyboard】 Shot 1 | 0—4 seconds Room A side medium shot, mirror and female lead in frame simultaneously. The female lead sits sideways on the bed edge, slowly removes one silver earring, then takes out the hairpin, long hair naturally falls behind her shoulders. Shoulders relaxed and sunk, neck elongated, gaze half-lowered. Earring and hairpin placed successively into the shallow dish on the dressing table. Hard cut to Room B medium shot: The Pharaoh presses against the hole observing, the Mechanical Gentleman lightly pats his shoulder urging him to switch positions, the Mouth Monster stretches its neck, the Lobster Monster retracts its pincers waiting. Only the Pharaoh has one eye at the hole, the four monsters cannot possibly see simultaneously. Shot 2 | 4—7 seconds Hard cut, Room A normal side full-body medium shot. The female lead supports herself on the bed with one hand, unties the ankle strap with the other, removes high heels in sequence, places them side by side beside the bed, feet slip into the pre-prepared flat slippers. When she raises her eyes, she discovers a rotating eye in the hole from the mirror surface. Movement pauses half a beat. She does not turn her head, only reaches out to gently turn the lamp shade, directing more light toward the shared wall, the eye in the hole thus clearly revealed. The female lead says in a low voice in Chongqing dialect: "Line up, one at a time." Shot 3 | 7—10 seconds Hard cut, Room B hole and four monsters medium shot. The Pharaoh jerks his head back, crown bumps the Mechanical Gentleman's bowler hat. The Mechanical Gentleman steadies his hat, the Mouth Monster dodges backward, the Lobster Monster hastily retracts pincers, the four's movements offset. Hard cut to Room A medium-wide shot: The female lead rises, picks up the dark gray nightrobe beside the screen, walks behind the screen. She leaves the mirror fully wearing the black dress. The real bed edge in the mirror is now empty. In the mirror, she still sits there, spine motionless. Shot 4 | 10—13 seconds Hard cut, Room B wall hole subjective POV. The Pharaoh can't resist leaning close again. Through the hole, the oblique mirror is still visible. The woman in the mirror slowly raises her head, gaze accurately toward the hole. She stands, walks out of the mirror's visible range. Cut to a Room A low-angle insert shot: The lower part of the mirror surface slightly concaves, a pale foot lands on the real floor, black dress hem falls accordingly. Then footsteps moving along the shared wall, direction toward the hole. The real female lead is not in this frame. Music stops, only light slow slipper-like footsteps remain. Shot 5 | 13—16 seconds Hard cut, Room B hole close-up. The originally warm-colored room inside the hole is suddenly completely filled by a pale eye. The eye stares at the Pharaoh through the fifteen-centimeter wall hole, eyelid slowly closes once. The Pharaoh jerks back, back slams into the Mechanical Gentleman. The Mechanical Gentleman disbelieves, sidesteps to replace him at the hole, presses his mechanical eye close to observe. A slender pale finger extends from the hole, lightly taps the mechanical eye's metal casing: "Ding." The Mechanical Gentleman's whole body freezes. The finger retracts along the original hole, only the eye remains in the hole again. Shot 6 | 16—20 seconds Hard cut, Room B doorway wide shot. The male lead carrying noodles passes by, hears the collision sound, lightly pushes the ajar door with his shoulder, pushes it open and enters. He does not kick the door, does not use the noodle bowl to hit the door panel. The four monsters immediately retreat from the wall hole, stand in a row, trying to block the shared wall with their bodies. The male lead looks at them, then at the hole in the wall, asks: "What are you doing here?" The Pharaoh straightens up, but his gaze keeps darting behind him: "Checking if the wall's cracked." Just as the words land, the slender finger extends from the hole again, slowly scrapes along the plaster edge. The four monsters hear the sound, shoulders tense simultaneously, but dare not turn around. Shot 7 | 20—24 seconds Hard cut, public corridor side medium shot. The real female lead has already put on the dark gray nightrobe, still fully wearing the black dress underneath, wearing flat slippers, exits Room A. She walks along the corridor to Room B's doorway, stops outside the door frame. Hard cut to Room B interior toward doorway medium-wide shot: The female lead holds the door frame, body slightly angled, shoulders relaxed and sunk, half-lowered eyelids, coldly sweeps across the four monsters: "Looks good?" The four monsters first look at the female lead, then instinctively look at the shared wall, realizing the same face is in both directions. The Pharaoh shakes his head first, the other three successively follow. Shot 8 | 24—27 seconds Hard cut, four monsters and wall hole in frame side medium shot. From Room A's wall hole comes a voice identical to the female lead's, but closer and slower: "I haven't seen enough yet." The real female lead at the doorway's mouth never moves. Her gaze for the first time shifts from the four monsters to the hole, confirming there is indeed something else inside. The four monsters slowly turn around. The pale eye in the hole rotates, stares at them in sequence. The Pharaoh takes a small step back, the Mechanical Gentleman reaches out to steady himself on the bedpost behind, the Lobster Monster shields its twin pincers in front of its face. Shot 9 | 27—30 seconds Hard cut, Room B wide full shot, doorway, wall hole and tall bed all visible simultaneously. The male lead holding the bowl steps half a pace aside toward the door, looks at the four monsters and says: "Now it's her turn to watch you." The Mouth Monster is closest to the bed, bends down and squeezes under the bed first. The Mechanical Gentleman follows closely, just half his body in, the Lobster Monster also tries to squeeze in, but its horizontal pincers jam outside the bed edge, blocking the Mechanical Gentleman's retreat. The Pharaoh arrives last, can only crouch at the bed's foot holding his crown, discovers there is completely nowhere to hide. Under the bed the Mouth Monster tries to shrink deeper, but its enormous head protrudes from the other side, still facing the wall hole. The eye in the wall hole slowly turns toward it. The real female lead stands at the doorway, shoulders relaxed and slightly angled, the corner of her mouth finally shows a trace of restrained smile. The male lead lifts a chopstick-full of noodles. A light plucked string note, hard cut to end. 【Dialogue & Audio】 All clear dialogue is natural Chongqing dialect: Female lead: "Line up, one at a time." Male lead: "What are you doing here?" Pharaoh: "Checking if the wall's cracked." Female lead: "Looks good?" Mirror double: "I haven't seen enough yet." Male lead: "Now it's her turn to watch you." The real female lead's tone is low, steady, ending with cold scrutiny. The ghost uses the same timbre, but the voice is close to the hole, breath slower; when speaking, the real female lead must not move her mouth. The male lead's tone is ordinary, no deliberate comedy. The Pharaoh pauses briefly when making excuses, cannot be exaggeratedly stuttering from the start. Ambient sound: Rain on window, earring and hairpin landing in shallow dish, shoe buckle soft clicks, old bed wooden frame, four monsters' clothing and armor friction, fingernail scraping plaster, crisp metallic sound of mechanical eye being lightly tapped. No frivolous music at the opening. After the mirror image acts independently, use extremely low sustained strings; leave silence when the ghost eye appears. Only after the four monsters hide under the bed add short comedic plucked strings. No narration, no subtitles, no additional clear dialogue. 【Continuity & Strict Constraints】 Only one real female lead, only one mirror double. The real female lead wears the dark gray nightrobe at the end, the double still in black dress, the two clearly distinguished. The female lead is fully clothed throughout, no undressing, nudity or changing process behind the screen shown. Voyeuristic POV only used to establish the four monsters' mistaken behavior and horror reversal, no explicit body composition used. Mirror angle fixed at the opening, must simultaneously support "hole POV sees female lead's reflection" and "female lead discovers eye in hole." After the female lead leaves, the real bed edge is empty, only then does the mirror double act independently. After the ghost enters Room A from the mirror, it walks along the shared wall to the hole's edge; does not directly appear in Room B, does not teleport beside the real female lead. The three-centimeter hole can only accommodate one slender finger, cannot pass a whole hand or whole face. The peeping eye, ghost eye and finger must appear in alternation, not penetrate each other. The four monsters only stand in a row after retreating from the hole, do not let broad bodies cram into the same position simultaneously. The male lead enters carrying the bowl, pushes the door with his shoulder, both hands' tasks are reasonable. The real female lead enters Room B's doorway along the corridor, movement route is clear. Space under the bed is limited, final comedy comes from scrambling to hide in sequence, pincers getting stuck and the enormous head not fitting, do not let the four monsters all vanish instantly. The female lead's siren-like posture follows real standing, shoe removal, walking and door-holding movements, no forced maintained crossed stance. No blood, no body tearing, no lasers, no magical particles, no sudden face flashes or unrelated monsters.
Create a high-quality 3D Pixar-style animated comedy video in vertical 9:16 format. Scene 1 – Parents Watching a Video: A beautiful young couple sitting together at a dinner table in a warm, elegant restaurant. The woman has long, wavy dark brown hair and wears a white top. The man has dark curly hair and wears a cream-colored shirt with brown suspenders. They are looking at a smartphone together, smiling with curiosity. Warm cinematic lighting, realistic facial expressions, detailed food and restaurant background. Scene 2 – Magical Pink Baby Playground: Cut to an adorable chubby baby with curly dark brown hair, huge sparkling blue eyes, and a cute white onesie. The baby is sitting in a magical pink fantasy playground filled with glowing stars, pink balloons, heart-shaped decorations, soft pink cushions, and a round trampoline. Scene 3 – Funny Baby Performance: The baby starts bouncing energetically on the trampoline, performing hilarious jumps, playful flips, and adorable acrobatic movements. Show a funny upside-down moment followed by the baby floating playfully in the air. Keep the movements smooth, exaggerated, and entertaining. Scene 4 – Parents Laughing: Cut back to the couple watching the phone. The woman bursts into uncontrollable laughter, holding her belly, while the man laughs beside her. Their facial expressions should look natural and genuinely amused. Scene 5 – Baby Close-Up: Show the baby flying playfully toward the camera with arms open, followed by an adorable close-up of the baby's big blue eyes and joyful smile. Final Scene: The baby sits on the pink trampoline, laughing happily, while glowing stars sparkle in the background. End with a funny, heartwarming expression. Visual Style: Ultra-detailed 3D Pixar-inspired animation, adorable character design, cinematic camera movements, expressive facial animation, soft global illumination, vibrant pink color palette, realistic fabric textures, smooth physics, professional animation quality. Important: Maintain consistent character appearance, facial features, hairstyle, clothing, and proportions throughout every scene.
Create an ultra-realistic, futuristic kitchen appliance commercial in vertical 9:16 format. Scene 1: A beautiful modern smart kitchen with soft natural daylight. A premium white and champagne-gold automatic food-making machine sits on a clean marble countertop. Its transparent front chamber opens smoothly, revealing a sleek futuristic interior with subtle cyan LED lighting. Scene 2: Fresh raw fish fillets are placed inside the transparent chamber. The machine automatically closes, and water droplets gently spray around the fish as an advanced cleaning process begins. Scene 3: Futuristic blue laser-like scanning lights move across the fish fillets. A magical transformation begins as golden crispy coating particles appear and surround the fish. Scene 4: The fish fillets gradually transform into perfectly golden, crunchy fried fish fingers inside the machine. Show realistic frying bubbles, steam, crispy textures, and appetizing golden colors. Scene 5: The machine automatically dispenses freshly cooked golden fish fingers into its bottom serving tray. Final Scene: A beautifully plated serving of crispy golden fish fingers appears on a white ceramic plate with lemon wedges, fresh herbs, and dipping sauces. The smart appliance stands elegantly in the background. Visual Style: Premium futuristic product commercial, photorealistic 3D animation, luxury kitchen aesthetics, cinematic lighting, smooth camera movements, realistic food textures, detailed reflections, elegant transitions, 4K quality. Camera: Smooth close-ups, slow-motion food transformation, cinematic product shots. No text, no subtitles, no watermark, no distorted food, no deformed appliance.
Preserve image’s exact identity, face, hairstyle, skin tone, body proportions, and reference outfit throughout. Authentic early-2000s Southeast Asian/Korean home-video atmosphere. Location: Simple residential house terrace with concrete steps, worn yoga mat, low boundary wall, potted plants, tangled overhead wires, soft morning light. No commercial activity. - 0–3s: Chaotic handheld camera swings, thumb briefly covers lens. Friend whispers, “Is it recording?” - 3–6s: She sits cross-legged on the mat, closes eyes, breathes calmly. - 6–9s: Raises arms and slowly folds forward into a stretch. - 9–12s: Notices camera, gives a shy finger wave and smile. - 12–16s: Moves into downward dog, naturally adjusting her heels. - 16–22s: Two curious kids peek over the wall, hide, then reappear giggling and waving. She laughs and waves back. - 22–26s: Rolls up mat and sits on the terrace steps. - 26–30s: Holds hot tea, blows on it, takes a sip, smiles directly at camera. Abrupt cut to black. Visual style: Extremely raw, shaky handheld early-2000s DV camcorder footage. Unplanned framing, accidental cropping, autofocus hunting, exposure pumping, motion blur, rolling shutter, digital noise, compression artifacts, faded desaturated colors, washed-out highlights. Imperfect amateur home-video realism. Audio: Natural ambience only—birds, breeze, distant motorcycles, camera handling, whispered conversation, laughter, children giggling, breathing, mat movement and tea glass clink. No music. Strictly avoid: Posing, commercial aesthetics, cinematic glamour, stabilization, perfect composition, modern color grading.
Ultra-realistic Korean K-Drama action thriller featuring Hollywood-level stunt choreography, brutal-looking close-range combat, explosive counterattacks, intense emotional expressions, realistic body physics, powerful impacts, dynamic handheld cinematography, rapid tracking shots, dramatic whip pans, and 120 FPS slow-motion impact inserts. Professional cinematic lighting, realistic sound design, grounded martial arts choreography, natural human movement, and physically believable collisions. Every fight sequence must feel intense, aggressive, unpredictable, and visually spectacular without supernatural abilities or cartoon physics. CHARACTERS FEMALE LEAD — @[Image 1]\(image_1) A beautiful 23-year-old Korean college student. Maintain her exact facial identity, hairstyle, clothing, body proportions, and overall appearance from the reference image throughout the entire video. She initially appears quiet, innocent, and vulnerable, but possesses exceptional martial arts skills, lightning-fast reflexes, extraordinary precision, and controlled aggression. MAIN BULLY — @[Image 2]\(image_2) An arrogant 24-year-old Korean male college student with a tall, muscular physique, messy platinum-blonde hair, a black leather jacket, white shirt, dark jeans, and black boots. Extremely aggressive personality, cruel facial expressions, intimidating body language, and explosive anger. BULLY 2 — @[Image 3]\(image_3) A tall, athletic Korean male student with short black hair, wearing a grey school blazer and black trousers. Aggressive, confident, and physically intimidating. BULLY 3 — @[Image 4]\(image_4) A muscular Korean male student with brown hair, wearing a black varsity jacket and dark pants. Violent, impulsive, and aggressive. SCENE 1 (0–4 SEC) — THE PUBLIC HUMILIATION A long corridor inside a modern Korean college. Metallic blue lockers line both sides of the hallway. Students walk through the corridor under bright overhead lighting. The female lead @[Image 1]\(image_1) stands beside her open locker, quietly taking out her books. Suddenly, the platinum-blonde bully @[Image 2]\(image_2) aggressively approaches her. BOOOOM! He violently slams her locker door shut with his hand. The girl instinctively steps backward, visibly startled. Without warning, the bully snatches the books from her hands and aggressively throws them into the air. Books and notebooks scatter across the polished corridor floor. The bully shouts angrily in English: "HEY! PICK THEM UP!" His two friends @[Image 3]\(image_3) and @[Image 4]\(image_4) burst into mocking laughter. The camera rapidly pushes toward the girl's face, capturing her shocked expression in an extreme close-up. Camera: Aggressive handheld movement, fast push-in, dramatic facial close-up. SCENE 2 (4–8 SEC) — THE FIRST ATTACK The girl quietly kneels to collect her scattered books. Suddenly, the main bully raises his right leg and launches a powerful kick toward her shoulder. WHOOOOSH! With lightning-fast reflexes, the girl instantly raises her left forearm and blocks the incoming kick. THUUUD! A brief 120 FPS slow-motion insert captures the precise moment of impact. The bully's expression immediately changes from arrogance to shock. Before he can recover, the girl explosively rises to her feet. She controls his extended leg, pivots sharply, and executes a rapid rotational sweep that destroys his balance. BAAAAM! The bully crashes sideways onto the floor beside the lockers. The metallic locker doors shake violently from the nearby collision. The camera follows his fall with a fast downward tracking movement. Camera: Extreme close-up, slow-motion kick block, explosive tracking shot. SCENE 3 (8–13 SEC) — THE LOCKER DOMINO Bully 2 angrily screams: "YOU'RE DEAD!" He charges directly toward the girl at full speed and throws a powerful right-hand punch. The girl instantly moves her head sideways, narrowly avoiding the strike. Without hesitation, she counters with a rapid body punch toward his ribs. THUD! She immediately follows with a powerful palm strike to his shoulder. THUD! The bully stumbles backward. The girl plants her supporting foot, rotates her hips, and delivers a lightning-fast spinning back kick. BOOOOOOM! The kick drives him backward into the metallic blue lockers. Three adjacent locker doors swing open from the vibration and collide against one another in rapid succession. CLANG! CLANG! CLANG! The metallic impacts create a spectacular domino-like chain reaction. The camera captures the entire collision in a low-angle lateral tracking shot. Camera: Rapid sideways tracking, low-angle impact framing, realistic metallic vibration. SCENE 4 (13–19 SEC) — TWO AGAINST ONE The main bully struggles back to his feet. At the same moment, Bully 3 charges forward. Both attackers rush toward the girl simultaneously. The main bully throws a violent left hook. The girl instantly ducks beneath the punch. Bully 3 launches a powerful front kick. She pivots sharply to the side. The kick misses its intended target, causing Bully 3 to collide awkwardly with his own friend. The girl immediately launches an explosive counterattack. Rapid combat choreography: A lightning-fast defensive block redirects an incoming strike. A powerful right palm strike forces the first attacker backward. A rapid spinning kick interrupts the second attacker's forward momentum. A low sweeping takedown knocks one bully off balance. A close-range shoulder throw sends the remaining attacker crashing onto the floor. BAM! THUD! CRASH! One bully collapses beside the lockers. The other slides across the polished corridor floor. The camera continuously circles the female lead, following every block, pivot, strike, and throw without unnecessary pauses. All movements must connect naturally, with believable momentum, realistic footwor