AI Video Prompts for Zoom Shots

Crash zooms that land like a punch, slow push-ins that build dread, snap zooms for editorial rhythm, pull-back reveals that show how small a figure is in a world — these Seedance 2.0 zoom prompts encode the full language of focal-length movement for AI video.

Zoom direction carries narrative intention. A zoom toward a subject signals increasing attention, threat, or revelation; a zoom away signals the growth of context, the smallness of a figure in a space, or the release of dramatic pressure. Both moves change focal length rather than moving the camera body through space, and that distinction from the dolly or tracking shot is the most important concept in zoom prompting. When you write "zoom" in a Seedance prompt, the model changes focal length, not camera position. That means the background compresses toward the subject as the focal length increases — the world behind the subject seems to approach as the subject is magnified. Compare this to a "push-in" or "dolly" instruction, which moves the camera body forward: the spatial relationship between subject and background remains proportional, only the distance changes. Knowing which one you want, and naming it correctly, is the difference between a zoom effect and a dolly effect — and they produce entirely different reads. The crash zoom is the most aggressive form: a sudden, violent jump in focal length that takes one to three frames to complete. It originated in Sergio Leone's spaghetti westerns, used to punctuate the instant of recognition between two figures before the draw. The crash zoom is not a camera style — it is a single punctuation mark, deployed at one specific beat: the moment of shock, the unexpected reveal, the comedic arrival. Prompts that use crash zoom correctly name the beat that triggers it and the starting distance — "medium establishing shot that crash-zooms to an extreme close-up on the face the instant the door opens." The rule: crash zoom once, at the exact right moment, in a clip that has been building to it. Used as ambient camera style, it produces disorienting footage with no payoff. The slow suspense zoom is the psychological opposite. Over five to ten seconds — barely perceptible at the start — the camera commits to a subject that the audience already suspects is the center of something. The technique that makes it land is placement: the slow zoom must begin after the tension peak, not before it. If the zoom starts when a character first enters the room, it reads as establishing camera; if it starts after the character stops and their face changes, it reads as the camera catching what they just understood. Prompts for slow suspense zoom name three things: the subject the camera is committing to, the speed register (barely perceptible, three-second push, sustained five-second push), and the timing relative to the scene beat — when in the action the zoom begins. The snap zoom sits between the crash and the slow push: a brisk focal-length shift over one to two seconds, used in two distinct contexts. In editorial cutting, it functions as a transition — the snap to a new focal length reads as a hard cut within a continuous shot. In social media and vlog content, it has become a genre convention: the snap zoom to a punchline, to a reaction, to an unexpected arrival. Prompts for snap zoom name the speed register, the direction (in or out), and the editorial purpose. The pull-back zoom — a slow or moderate zoom out from a subject — reveals context around a figure and is structurally the richest move for endings. It begins on a detail or a face and ends on the full space, showing how isolated, how small, or how situated the subject is within the world. The emotional register ranges from melancholic (the figure growing smaller as the city expands behind them) to triumphant (the narrow focus opening into a wide landscape at the moment of resolution). Prompts for pull-back zoom name the starting frame, the ending frame, and the emotional register the movement is meant to carry. The dolly zoom — the Vertigo effect, the Hitchcock zoom — is the most technically complex and cinematically powerful technique, and it is available in AI video when named precisely. The mechanism: the camera body moves backward while the focal length simultaneously increases, at rates calibrated so the subject's apparent size stays constant. The result is that the subject holds still while the background dramatically changes scale — the space behind them seems to stretch or collapse, expressing vertigo, dread, or the ground dropping away. To prompt the dolly zoom in Seedance, both movements must be named simultaneously: "the camera dollies backward while simultaneously zooming in, maintaining the subject's apparent size — the background scale shifts dramatically behind them." Without naming both the camera body movement and the compensating focal length change, Seedance executes only one — and the Vertigo effect requires both at once. Across all zoom types, the structural rule is the same: name the zoom direction, the speed, the trigger beat, and the starting and ending frames. Zoom prompts that skip any of these four elements produce a generic focal-length shift without dramatic purpose — the camera moving for its own sake rather than the story's.

Seedance 2.0 AI video prompt: 0-5 seconds
Camera Work: Low-angle shot, showcasing the immense body of the drag

0-5 seconds Camera Work: Low-angle shot, showcasing the immense body of the dragon. Flames and magma splash around, and the dragon's scales glisten in the firelight, creating a tense and oppressive atmosphere. Action: The samurai stands firmly in the center of the cave, feet planted, holding the glowing katana, eyes locked on the dragon. 6-9 seconds Camera Work: The camera rapidly zooms in, capturing the dragon opening its massive mouth, letting out a deafening roar. Flames surge from the dragon’s mouth, the air vibrates, and the entire cavern fills with the overwhelming sound. Action: The dragon's roar stirs up a powerful gust of wind, causing the samurai's clothes to flap violently, but he remains steadfast and unmoved. 10-12 seconds Camera Work: The camera quickly pans upward as the dragon roars, capturing the samurai suddenly leaping into the air, swiftly raising his glowing katana. The motion is smooth and swift, the sword slicing through the air with a flash of light. Action: The samurai leaps towards the dragon's head, the tip of the katana precisely touching the dragon's forehead. The blade emits a dazzling glow, the light spreading from the sword’s tip, filling the entire frame, as if the air has frozen. 13-15 seconds Camera Work: The camera begins a 360-degree rotation, slowly circling the samurai and the dragon, emphasizing the moment of the sword’s contact with the dragon's head. The rotating camera highlights the light emanating from the sword, spreading outwards, bringing a sense of mysterious power. Action: As the sword touches the dragon’s head, the glow intensifies. The samurai and the dragon’s eyes slowly close, entering a deep resonance. The surrounding flames and smoke gradually subside, and the atmosphere becomes calm and mystical. The circular motion of the camera adds an epic feel, as if time has completely stopped.

Seedance 2.0 AI video prompt: Ultra-realistic cinematic video, 15s, 24fps.

A female dancer performs smooth hi

Ultra-realistic cinematic video, 15s, 24fps. A female dancer performs smooth hip-hop choreography (including a moonwalk). Her body motion is perfectly continuous and natural throughout. CAMERA: Dynamic handheld energy — swings, slight shake, crash zooms, fast push-ins, and subtle orbits synced to her movement. CORE: She stays centered, full body visible, dancing confidently with strong rhythm. CHAOS (HIGH SPEED): Every frame rapidly changes clothing, visual style, and environment. CLOTHING: New trendy outfit every frame (no repetition): streetwear, layered urban, edgy fashion, performance wear. STYLE (identity preserved): Rapid switching between: realistic, anime, line art, watercolor, comic, pastel, digital paint, sketch, collage, clay, low-poly, neon outline, glitch graphic, storybook, chalk, airbrush, silhouette. Occasional 2–3 style blends for 1 frame bursts. ENVIRONMENT (KEY): Background constantly transforms in real-time around her — NOT hard cuts. Environments include: subway, neon city (wet), white studio, brutalist concrete, rooftop sunset, warehouse haze, digital grid, marble hall, graffiti alley, desert, futuristic corridor, black void. TRANSITIONS (MOTION-DRIVEN): - camera whip → environment shifts - spin → 360° background morph - moonwalk → ground transforms under feet - crash zoom → new environment revealed - light flicker → environment rebuilds DEPTH: Real parallax (foreground / midground / background). No flat backgrounds. LIGHTING: Adapts to each environment (neon glow, studio soft light, industrial beams, digital emissive) and stays consistent on her body. CRITICAL: - dancer locked in position (no teleporting) - no body or face distortion - environment changes around her only - motion must stay perfectly continuous SYNC: All changes tightly follow beat, movement, and camera rhythm. VISUAL: Hyper-fast, energetic, but clean and readable. NO text, NO logos.

Seedance 2.0 AI video prompt: Cinematic 15-second luxury house tour composed of 15 rapid 1-second shots, each

Cinematic 15-second luxury house tour composed of 15 rapid 1-second shots, each cut cleanly with smooth visual continuity, ultra-realistic interior and exterior architecture, high-end modern luxury home, natural lighting, cinematic color grading, consistent style across all shots. Shot list sequence: 1. Exterior establishing shot of modern luxury villa with landscaping and warm sunlight 2. Slow push through grand entrance doorway 3. Wide living room with floor-to-ceiling windows and designer furniture 4. Close-up of marble coffee table with decor details 5. Pan across double-height ceiling and chandelier 6. Luxury kitchen island with stone countertop and soft daylight 7. Slow slide past premium appliances and cabinetry 8. Dining area with large table and elegant lighting 9. Bedroom wide shot with soft linen bedding and natural light 10. Walk-in closet with warm LED shelf lighting 11. Bathroom with freestanding tub and stone textures 12. Balcony or terrace with outdoor seating and view 13. Staircase with architectural lighting and glass railing 14. Home lounge / media room with ambient lighting 15. Final exterior sunset hero shot of the house Fast cinematic cuts, smooth motion in every shot, micro camera movement per shot (push, pan, slide), physically accurate lighting, realistic reflections, soft shadows, ultra-realistic materials, marble, glass, wood, metal. Consistent exposure, no flicker, stable geometry, real-world motion blur, shallow depth of field where appropriate, HDR, ultra high definition, film-quality luxury real estate cinematography.

Seedance 2.0 AI video prompt: Film stock: 35mm Kodak Vision3 500T, heavy organic film grain, high contrast.
Le

Film stock: 35mm Kodak Vision3 500T, heavy organic film grain, high contrast. Lens/Aperture: 35mm Anamorphic lens, f/2.8. Deep depth of field to see both characters clearly. Color Grade: "Saturated 90s Diner" palette. Warm nicotine yellows, bright red vinyl booths, and harsh fluorescent overheads. Camera Behavior: Slow, rhythmic "Shot/Reverse Shot" switching. Starting with a slow creeping zoom on the lead's face. Atmosphere: A half-empty, sun-drenched diner. Dust motes floating in the light. Tense, quiet, blue-collar grit. Audio: Immersive spatial sound design. The distant clinking of silverware, a coffee pot pouring. Dialogue lipsync: Character @ image1 leans in and says: "I'm gonna ask you one more time, and if you lie, God himself won't be able to find what's left of you." [IMAGE REFERENCES / LEGEND] @ image1: The lead enforcer. Maintain exact beard, dark sunglasses, and black headphones (worn around the neck for this scene). Keep exact same character, style, and lighting. Character 2: A skinny, sweating man in a cheap, rumpled grey suit sitting opposite him, trembling while holding a ceramic coffee mug. [TIMELINE SECOND BY SECOND] 0-4s: [Medium Shot - Over the Shoulder] + [Focus on Character @ image1] + [Action: He is calmly stirring a cup of black coffee with a silver spoon] + [SFX: Rhythmic metallic tink-tink-tink of the spoon against porcelain]. 4-8s: [Close-up] + [Dialogue lipsync: Character @ image1 stops stirring, looks up slowly, and delivers the line with a cold, terrifying calm] + [Physics: Small wisps of steam rising from the coffee cup]. 8-11s: [Reverse Shot] + [Focus on Character 2] + [Action: He visibly gulps, a bead of sweat rolling down his forehead, eyes darting nervously] + [SFX: Muffled sound of a waitress laughing in the far background]. 11-15s: [Low-angle Profile Shot] + [Character @ image1 slowly reaches into his pocket, pulling out a heavy, chrome-plated handgun and placing it quietly on the table next to his saucer] + [Lighting: Sunlight glints sharply off the chrome]. [STYLE & QUALITY BOOSTERS] Movie-level realistic facial features, no deformation, stable character consistency. Professional 90s crime film aesthetic. High-fidelity skin textures (pores, sweat).

Seedance 2.0 AI video prompt: Animate the provided first frame into a 15-second cinematic K-pop dance performa

Animate the provided first frame into a 15-second cinematic K-pop dance performance video in a 16:9 horizontal frame. Keep the same four fictional performers, same faces, same outfits, same hairstyles, same desert festival stage, same sunset lighting, same LED screens, and same black-pink-gold concert aesthetic throughout the entire video. STYLE: High-budget K-pop concert film, Coachella-inspired desert festival atmosphere, glossy music video cinematography, sunset backlight, pink and gold LED lighting, stage haze, pyrotechnic smoke, reflective black stage floor, blurred cheering crowd, cinematic camera movement, polished idol choreography. CHARACTER LOCK: There are exactly four female performers for the entire video. They are an original fictional K-pop-style group, not real celebrities and not existing idols. Keep each member visually consistent: Center member: long black wavy hair, black crystal crop top, pink metallic skirt, knee-high black boots. Left member: blonde high ponytail, black cropped jacket, gold chain belt, fitted black cargo pants. Right member: rose-pink hair, sleeveless stage top, black shorts with metallic accents, high boots. Rear member: dark brown bob, pink-and-black jacket, silver accessories, athletic stance. Do not swap their faces, hairstyles, outfits, positions, or identities. TIMELINE: 0–1.5s: Start exactly from the first frame. The four performers hold a powerful diamond formation for a brief dramatic beat. Sunset light glows behind them, LED screens pulse softly in pink and gold, haze moves across the stage. 1.5–3s: The beat drops. All four snap their heads up, open their arms outward, and hit a sharp synchronized shoulder accent. Camera performs a slow low-angle push-in from the front. 3–4.5s: Center performer steps forward half a step while the other three angle outward. The group performs a clean arm sequence: wrist flick, elbow hit, shoulder pop, then a strong pose. Stage lights flash white-pink on the beat. 4.5–6s: The dancers slide laterally across the glossy stage in formation, then shift back to center. Hair, jackets, belts, and accessories move naturally. Keep full-body visibility so the footwork is readable. 6–7.5s: Medium close-up on the blonde performer as she turns sharply and performs a confident hand gesture near her face. The other members remain visible behind her, continuing smaller synchronized accents. 7.5–9s: Return to a wide shot. The group changes from diamond formation into a horizontal line, then performs a synchronized body roll followed by a diagonal arm cut. LED screens display abstract pink and gold light waves, no text. 9–10.5s: Camera moves into a smooth partial orbit around the group from front-left back to center. The performers execute fast but clean footwork, one step-cross sequence, and a sharp hip accent. Avoid chaotic motion blur. 10.5–12s: Chorus peak. Pyrotechnic sparks burst safely at both sides of the stage. The four performers hit a strong pose, then perform a synchronized hair flip and hand accent. Haze and spotlights create dramatic silhouettes. 12–13.5s: The group performs a coordinated jump or strong level change, landing perfectly together. Confetti begins falling lightly from above. The reflective stage floor catches their boots, legs, and pink-gold lights. 13.5–15s: Camera settles into a centered low-angle hero shot. The performers return to a tight diamond formation with the center member slightly forward. They finish in a powerful final pose facing the camera, sunset blazing behind them, crowd blurred in the background, LED panels glowing pink and gold, pyrotechnic smoke framing the stage. CAMERA: Use smooth concert cinematography: low-angle push-in, wide full-body choreography shots, one medium close-up, side tracking, partial orbit, and final low-angle hero shot. Keep choreography readable and synchronized. Avoid excessive cutting. MOTION: The choreography should feel fierce, glamorous, and professional: sharp arm hits, shoulder pops, body rolls, clean footwork, formation changes, hair flips, level change, and final pose. Movement should match an imaginary high-energy K-pop festival chorus around 125–135 BPM. LIGHTING: Golden sunset backlight, hot pink LED panels, white concert spotlights, warm rim light on hair, subtle haze, glossy black stage reflections, pink and gold highlights. Lighting should be dramatic but faces must remain clear. CONTINUITY RULES: Exactly four performers only. No extra backup dancers. No missing members. No identity swapping. No outfit changes. No hairstyle changes. No real celebrity likeness. No readable logos or text. Keep the desert festival stage and sunset environment consistent from start to finish.

Seedance 2.0 AI video prompt: Character Description
Young woman in her early 20s with the exact same facial fe

Character Description Young woman in her early 20s with the exact same facial features as the reference image, realistic appearance, soft oval face, fair warm skin tone, expressive almond-shaped hazel eyes, naturally thick brows, delicate nose, soft pink lips, gentle elegant smile, realistic skin texture, loosely tied brunette hair with soft face-framing strands. Wearing a stylish red-and-black football jersey layered over a gray oversized hoodie, casual streetwear vibe, slim build, slightly shy personality, authentic candid facial expressions, seated among football supporters inside a packed stadium. No celebrity look, no model styling, completely ordinary realistic sports fan energy. Updated Version (Football Stadium + English Commentary) Second 1 Wide shot of a massive professional indoor football stadium in South Korea. A giant hanging jumbotron dominates the center of the arena. Bright stadium floodlights, realistic crowd movement, subtle audience chatter, distant referee whistles, and authentic football match atmosphere. The jumbotron displays a young female audience member wearing a red-and-black football jersey layered over a gray oversized hoodie, casually sitting in the stands. Faint English sports commentary can be heard naturally in the background: “Looks like we’ve found one of the fans enjoying tonight’s match…” Second 2 Camera remains focused on the jumbotron screen. Subtle Korean live-TV graphics appear naturally in the corner: minimalist Korean broadcast interface, translucent scoreboard overlay, small Korean text, tiny red LIVE icon. The young woman still does not realize she is being shown on the giant stadium screen. English commentator: “She has absolutely no idea she’s on camera right now.” Second 3 People sitting nearby begin noticing her on the screen. Some spectators laugh softly and point toward her. The young woman starts looking around in confusion, trying to understand what is happening. English commentator chuckles lightly: “And there it is… realization is slowly kicking in.” Second 4 The broadcast camera smoothly tilts downward from the hanging jumbotron toward the real audience seating area. Realistic long-lens sports camera movement with subtle handheld micro-shake from a live TV operator. Crowd noise swells naturally. Second 5 The camera finally locates the young woman in the crowd. Autofocus briefly adjusts and locks onto her face. Slow zoom-in begins naturally like a real football broadcast reaction cam. English commentator: “We found her. Front row of embarrassment tonight.” Second 6 The zoom becomes tighter and slightly faster. The young woman suddenly realizes the live camera is focused on her and that her face is being displayed on the stadium screen. Her expression instantly becomes awkward and embarrassed. Nearby fans laugh naturally. English commentator laughing softly: “There it is! She knows now.” Second 7 Medium close-up shot. She gives a nervous laugh, briefly looks away from the camera, shoulders slightly tense. Her hands move awkwardly as if unsure how to react. Nearby audience members smile and laugh naturally. English commentator: “You’ve got to wave now. Stadium rules.” Second 8 The camera zooms even closer in authentic live-TV fan-cam style. Background spectators become softly blurred due to telephoto lens compression and shallow depth of field. The young woman forces a shy smile while partially covering her face. Subtle compression artifacts and realistic sports-broadcast sharpness. Second 9 Tight realistic close-up. She avoids direct eye contact, laughs awkwardly, then gives a small embarrassed wave toward the camera. Crowd reactions remain subtle and natural. English commentator: “There we go! She survived the fan cam.” Second 10 The camera slightly shakes as if the live-TV operator is adjusting the shot while staying locked onto her awkward expression. Korean LIVE broadcast overlays remain visible. The moment feels spontaneous, candid, and completely authentic like a rea

Seedance 2.0 AI video prompt: [CINEMATIC SETUP]

Film Style: Photorealistic 8K, 50mm anamorphic. Brass desk la

[CINEMATIC SETUP] Film Style: Photorealistic 8K, 50mm anamorphic. Brass desk lamp casting hard amber light from screen-left onto a walnut desk. Fast editorial montage — snap-zoom between globe surface and civilization vignettes. Camera: Locked on desk for globe shots. Snap-zoom push-in for vignettes, snap-zoom pull-out to return. [TIMELINE — MONTAGE] 0-1s: DESK. A detailed globe on a walnut desk under lamp light. A hand spins it. 1-3s: Globe stops on Middle East. Snap-zoom in — painted surface dissolves into photorealistic aerial of irrigated farmland, ziggurats, oxen plowing mud. Golden dust. SFX: wind, chanting. 3-5s: Snap back to desk. Hand flicks globe. Stops on Africa. Snap-zoom in — ground-level view of Djenne mud mosque, gold traders crossing a market square, indigo-robed horsemen. SFX: drums, chatter. 5-7s: Snap back. Spin. Stops on South America. Snap-zoom — Machu Picchu at sunrise, terraces descending into cloud forest, condors circling. SFX: mountain wind, distant flute. 7-9s: Snap back. Spin. Stops on Europe. Snap-zoom — medieval battlefield, cavalry charging across mud, siege towers rolling toward a fortress, catapult arcs. SFX: hooves, metal clash. 9-11s: Snap back. Spin faster. Stops on East Asia. Snap-zoom — coastal trading port, wooden junks in harbor, silk bales on a wharf, fireworks above a pagoda. SFX: firecrackers, bells. 11-13s: Snap back. Hard spin. Stops on North America. Snap-zoom — Manhattan assembling in timelapse, steel girders rising, bridges spanning, electric lights flooding the grid. SFX: riveting, electric hum. 13-15s: Snap back. Hand lifts globe off stand. Lamp flickers off. Globe dissolves into photorealistic Earth from orbit — city lights blazing, a rocket trail arcing into space. Hand opens, Earth floats free. SFX: silence, single low tone. [QUALITY] Photorealistic 8K. Desk consistent as anchor between vignettes. Snap-zoom bridges painted globe to real landscape through shared geography. Each vignette 3-4 details, one primary motion. No text overlays. Consistent lamp color on desk returns.

Seedance 2.0 AI video prompt: Cinematic absurd fantasy time-slip comedy featuring an orange cat in an exaggera

Cinematic absurd fantasy time-slip comedy featuring an orange cat in an exaggerated soap-opera style. Ultra-realistic visuals with extreme melodrama and explosive comedic timing. Character: Orange cat: dramatic, jumpy, overreacts instantly. Wearing a slightly wrinkled gray suit. White cat: peacefully asleep in soft pink pajamas. Guinea pig: tiny, wet, aggressively affectionate. Gorillas: primitive, expressive, laugh uncontrollably. Environment: 1950s Japanese tatami room. Dim warm light. Covered mirror in the corner, subtly shaking. Camera: Extreme close-ups → slow suspense push → sudden zoom → violent pull → rapid absurd cuts → chaotic comedy framing. Sequence: 0–2s: white cat sleeping. SFX (whisper): “Cat master… wake up…” 2–4s: orange cat eyes snap open. Orange Cat: “...Nope.” 4–6s: hears it again, turns slowly. SFX: “Cat master…” Orange Cat: “Who said that?!” 6–8s: tiptoes dramatically. Checks white cat. Orange Cat: “Stay asleep. Please.” 8–10s: approaches mirror. SFX (closer): “Come closer…” Orange Cat: “This is a bad idea…” 10–11.5s: lifts cloth. FLASH. 11.5–12.5s: sucked in violently. Orange Cat: “NO NO NO—!!” 12.5–13.5s: time vortex chaos. Orange Cat: “TOO MUCH!!” 13.5–14.2s: lands in snowy hot spring (Mount Fuji). SFX: SPLASH!! Orange Cat: “I regret everything.” 14.2–14.7s: looks up, trembling. Orange Cat: “Where am I…” 14.7–15s: guinea pig pops up and hugs his face. SFX: SQUEAK! Orange Cat (muffled): “WHY?!” Gorillas explode laughing. SFX: HAHAHAHAHA!! Orange Cat (to camera): “HELP ME!!” Style: Ultra-realistic with absurd fantasy, fast pacing, dramatic zooms, exaggerated reactions, punchy comedic beats. Final: Consistent characters, smooth transitions, no glitches, maximum chaos, strong slapstick payoff.

Seedance 2.0 AI video prompt: Create a 15-second continuous one-shot 2D animation with no hard cuts.

The subj

Create a 15-second continuous one-shot 2D animation with no hard cuts. The subject is [MAIN CHARACTER]. Do not depend on any specific character reference. Preserve the character as one recognizable main subject, but treat the body, hair, clothing, and surrounding world as flexible graphic material that can continuously melt, stretch, smear, spiral, flatten, dissolve, and reform. The main visual event is not ordinary movement. The main visual event is the character’s body transforming. The character moves through the scene by fluid transformation rather than walking, running, or drifting normally. Whenever the character travels across the screen, the body should melt forward, smear, pour, stretch, ribbon, and flow into the next pose while remaining one connected continuous form. The transition between poses must itself be visible as a transformation event. The body should frequently pass through these states: 1. a readable full-body pose 2. a partially melted body 3. a face-dominant state where the body flows away into ribbons and liquid bands 4. a spiral, helix, corkscrew, or whirlpool-like deformation 5. the body merging into the abstract background 6. re-formation into the same recognizable character Major body transformation must happen every 1 to 2 seconds. Do not keep the character stable for long. The figure should repeatedly break apart, flow, twist, collapse, stretch, and reconnect. The character must remain one single continuous connected form. Do not create extra human silhouettes, clones, shadow figures, duplicate bodies, detached outline people, or additional characters. All melting, spiraling, unraveling, and flowing deformation must stay physically connected to the original character. The face should remain the emotional anchor whenever visible. Even when the body is heavily melted, spiraled, or abstracted, the face should stay readable at key moments. Expressions should remain positive, bright, and uplifting: joy, surprise, laughter, excitement, emotional wonder, soft inspiration, and delighted curiosity. Do not use crying expressions, tears, sadness, grief, depression, watery eyes, sorrowful faces, or downcast emotional states. This must be purely flat 2D animation. No 3D rendering. No CGI look. No volumetric lighting. No realistic depth shading. No sculpted 3D form. No glossy 3D surface treatment. No pseudo-3D camera feeling. Use flat graphic animation, solid fills, clean edges, minimal shading, hand-drawn energy, crisp silhouettes, and planar motion. The background must not be passive. The background should actively flow, deform, ripple, unwrap, shift patterns, and merge with the character. The character and background should feel like one living 2D pattern system. Use a rich abstract visual field made of: organic blobs, wavy lines, curly ribbons, liquid bands, spiral trails, dots, sparkles, arrows, stars, crescent shapes, geometric particles, soft cutout shapes, floating patches, and evolving flat patterns. These motifs should be abundant and constantly active. They should drift, bounce, orbit, stretch, liquefy, collide, multiply, merge, and reconnect with the character’s hair, face framing, clothes, arms, torso, and body flow. Do not create emoji-like faces, emoticons, kaomoji, smiley icons, decorative cartoon faces, face stickers, or symbolic face motifs in the background or effects. Only the main character may have a face. All surrounding motifs must remain abstract, botanical, geometric, liquid-like, or graphic. Color direction: Use [COLOR PALETTE]. Keep the image flat and graphic: solid fills, clean edges, minimal shading, no realism, no heavy gradients, no volumetric rendering. Motion direction: Sync the motion tightly to [BPM]. Use buoyant beat hits, elastic rebounds, upward floating surges, rhythmic liquid wobble, spiral deformation, body-smear trails, rapid morphing, ribbon-like unraveling, connected paint-flow distortion, soft glitch pulses, geometric pop accents, and dramatic re-formation. The screen should stay visually rich at all times. Do not leave large empty static moments. Do not reduce the animation to simple drifting. New flowing shapes, connected ribbons, liquefied body parts, merging motifs, recombining graphic elements, and evolving surface patterns should keep appearing throughout the shot. Camera: One continuous shot only. The camera may push in, pull back, drift sideways, roll, tilt, orbit, and dynamically reframe the subject. The camera should follow and amplify the transformation, but the world must still read as flat 2D animation, not 3D space. Time structure: 0:00 - 0:01.5 Begin with one readable full-body pose of the main character floating inside an abstract 2D pattern world. Immediately introduce motion density. The camera pushes in while drifting slightly. Hair, clothing edges, sleeves, hands, feet, or lower body begin to melt outward into ribbons and liquid bands. 0:01.5 - 0:03.0 The character moves across the frame by melting forward rather than stepping. The torso pours into the next position as if pulled by elastic liquid force. Arms and legs smear into connected streaks. The face stays readable while the body trails behind as one continuous connected form. Background patterns are pulled along with the body movement. 0:03.0 - 0:04.5 Shift into a face-dominant transformation. The face becomes the emotional anchor while hair, torso, limbs, and clothing dissolve into swirling abstract bands and attached blobs. The background folds into the body flow. Expression shifts to surprise, delight, or wonder. 0:04.5 - 0:06.0 The body briefly reforms, then twists into a helix or corkscrew deformation. The whole figure spirals like a ribbon screw while remaining one connected body. Clothes, hair, limbs, and abstract motifs interweave. Use dense spiral trails, liquid bands, and rhythmic graphic accents. 0:06.0 - 0:07.5 The figure collapses into a flowing abstract mass. Only parts of the face or hair silhouette remain visible. The body spreads sideways like liquid cut-paper, then rebounds upward with elastic energy. The background becomes especially active and merges into the collapsing body. 0:07.5 - 0:09.0 The character reforms into a recognizable figure for a short moment, then immediately melts into travel again. The body pours diagonally through space. One arm unravels into ribbons, the torso stretches, the legs flatten into graphic bands, then everything reconnects. 0:09.0 - 0:10.5 Enter another face-dominant transformation phase. The face stays centered while the rest of the body streams outward into liquid trails, spiral loops, and abstract cutout shapes. Motifs orbit, collide, and fuse back into the body mass. No duplicate bodies or detached human copies. 0:10.5 - 0:12.0 The body twists again into a major spiral or whirlpool state. Hair, clothing, torso, arms, and legs become one connected flow. Background patterns wrap around the character and are absorbed into the deformation. Use strong rhythmic rebounds and organic glitch-like pulses. 0:12.0 - 0:13.5 Build toward a climax. The character and background nearly merge into one flat 2D living pattern system. The face appears, disappears, and reappears through ribbons, patches, liquid matter, stars, bands, and flowing shapes. The figure must not hold still. Major visible transformation continues every beat or two. 0:13.5 - 0:15.0 Final climax and re-formation. The whole screen reaches peak motion density as the face, hair, outfit, limbs, torso, and background flow together as one connected abstract system. Then the character reforms into one strong readable final frame with a bright, uplifting expression. The ending should feel like the peak of continuous transformation, not a calm stop. The final frame should still contain active flowing pattern energy around the re-formed character.

Seedance 2.0 AI video prompt: [CINEMATIC SETUP]

Film Style: Photorealistic 8K, 50mm anamorphic. Brass desk la

[CINEMATIC SETUP] Film Style: Photorealistic 8K, 50mm anamorphic. Brass desk lamp casting hard amber light from screen-left onto a walnut desk. Fast editorial montage — snap-zoom between globe surface and civilization vignettes. Camera: Locked on desk for globe shots. Snap-zoom push-in for vignettes, snap-zoom pull-out to return. [TIMELINE — MONTAGE] 0-1s: DESK. A detailed globe on a walnut desk under lamp light. A hand spins it. 1-3s: Globe stops on Middle East. Snap-zoom in — painted surface dissolves into photorealistic aerial of irrigated farmland, ziggurats, oxen plowing mud. Golden dust. SFX: wind, chanting. 3-5s: Snap back to desk. Hand flicks globe. Stops on Africa. Snap-zoom in — ground-level view of Djenne mud mosque, gold traders crossing a market square, indigo-robed horsemen. SFX: drums, chatter. 5-7s: Snap back. Spin. Stops on South America. Snap-zoom — Machu Picchu at sunrise, terraces descending into cloud forest, condors circling. SFX: mountain wind, distant flute. 7-9s: Snap back. Spin. Stops on Europe. Snap-zoom — medieval battlefield, cavalry charging across mud, siege towers rolling toward a fortress, catapult arcs. SFX: hooves, metal clash. 9-11s: Snap back. Spin faster. Stops on East Asia. Snap-zoom — coastal trading port, wooden junks in harbor, silk bales on a wharf, fireworks above a pagoda. SFX: firecrackers, bells. 11-13s: Snap back. Hard spin. Stops on North America. Snap-zoom — Manhattan assembling in timelapse, steel girders rising, bridges spanning, electric lights flooding the grid. SFX: riveting, electric hum. 13-15s: Snap back. Hand lifts globe off stand. Lamp flickers off. Globe dissolves into photorealistic Earth from orbit — city lights blazing, a rocket trail arcing into space. Hand opens, Earth floats free. SFX: silence, single low tone. [QUALITY] Photorealistic 8K. Desk consistent as anchor between vignettes. Snap-zoom bridges painted globe to real landscape through shared geography. Each vignette 3-4 details, one primary motion. No text overlays. Consistent lamp color on desk returns.

Seedance 2.0 AI video prompt: LOOK:
raw film camera footage, overcast rainy day (grey cloudy sky, steady moder

LOOK: raw film camera footage, overcast rainy day (grey cloudy sky, steady moderate rain with visible streaks — not a downpour, kept light enough to read faces — wet reflective streets and puddles, soft flat daylight). Same rain intensity in every shot. Natural light, NO color grading — looks like real unedited footage, not glossy, not graded. CAMERA: handheld shaky throughout, highly dynamic: quick zoom in/out, fast whip-pans, snap pushes timed to throws/catches/impacts; steadier only on tight close-ups. PERFORMANCE (distinct acting per character): - the Clown: rubbery, theatrical circus-clown physicality — bouncy springy movement, big sweeping arm flourishes, exaggerated lunges and recoils, elastic face (eyes popping wide, brows flying, mouth stretching huge), gleeful and unhinged, always over-the-top. - the Colonel: cranky old-boomer energy — stiffer, slower, heavier; dismissive scowls and eye-rolls, a sour set mouth, deliberate cane taps and curt hand waves, grumpy and unimpressed even when he attacks. Both read big and comedic but stay fully photoreal, never animated-looking. EMOTION = BODY (not labels): show every feeling through visible movement: eyes, brows, mouth, head tilt, hands, shoulders, stance. Keep both characters alive in every shot even when still: breathing, blinks, small weight shifts, rain landing on them. VOICE: lifelike voice (use this voice @[Audio 1](audio_1) for Colonel / use this voice @[Audio 2](audio_2) for Clown): natural breaths and rhythm, not flat TTS; the emotion of each line matches the action on screen (furious, taunting, straining, weak, smug). CHARACTERS: "the Colonel" (KFC — white suit, white goatee, glasses, cane) vs "the Clown" (McDonald's — red hair, yellow jumpsuit, red-white stripes); they duel by hurling food across the street — Colonel throws fried chicken, Clown throws burgers. @[Image 1](image_1) is the Colonel, @[Image 2](image_2) is the Clown ENVIRONMENT (identical every gen): rainy wide American downtown street: overcast grey sky, steady moderate rain (visible streaks, not a downpour), wet reflective asphalt, puddles; two fast-food restaurants on opposite corners (red-white vs yellow-red). Reference @[Image 3](image_3) is wide street + @[Image 4](image_4) is KFC Store, @[Image 5](image_5) is McDonald's Store STAGING (fixed): Colonel always at the KFC storefront, Clown always at the McDonald's storefront opposite, facing across the street — never together in the middle; each keeps his own store visible behind him. AUDIO: no music or score; only diegetic sound (steady rain, footsteps in puddles, food whooshes, wet impacts and splashes, street tone). TEXT: No on-screen text or subtitles. Shot 1: medium shot on the Colonel standing right in front of KFC store — he lifts one hand to catch it... the burger swerves past his palm and splats square into his face, then peels slowly down leaving a smear of sauce and leftovers. quick zoom in, his face begins to tremble — a vein bulging at his temple, jaw muscles twitching, nostrils flaring, one eye ticking, knuckles whitening as his fist crushes the cane — fighting with everything he has to hold his temper in. He drags one slow, shaking breath through gritted teeth. Then, low and quaking with barely-contained fury, right on the edge of erupting but forcing it down: "You brought a Happy Meal to a gunfire, boy." Shot 2: hard cut to high angle close up shot of the Clown doubled over laughing, head thrown back, slapping his knee, eyes squeezed to manic slits. He howls: "What about a cannon?" laugh and slapping his knee

Seedance 2.0 AI video prompt: TITLE: TOKYO NIGHTS — 15 SECOND HIGH-ENERGY JAPAN VLOG

STYLE

• vibrant nightli

TITLE: TOKYO NIGHTS — 15 SECOND HIGH-ENERGY JAPAN VLOG STYLE • vibrant nightlife vlog • rapid-fire editing • handheld camera • fast whip-pan transitions • energetic Japanese rock song throughout • no slow motion • authentic Japanese dialogue • colorful neon lighting • social, fun, chaotic energy • travel vlog aesthetic AUDIO Japanese rock song playing continuously. Crowd ambience. Laughter. Street sounds. Arcade sounds. Restaurant chatter. ━━━━━━━━━━━━━━━━━━━━ 0.0s - 2.0s HOOK SHOT Neon-lit Tokyo nightlife street. Two stunning Japanese women rush into frame from opposite sides, both speaking directly to camera simultaneously with huge smiles and excited energy. WOMAN #1 (Japanese): 「みんなー!今夜は最高に盛り上がるよー!」 WOMAN #2 (Japanese): 「準備はいい!?一緒に遊ぼう!」 Both laugh. Quick camera push-in. Japanese rock song drops immediately. WHIP PAN TRANSITION. ━━━━━━━━━━━━━━━━━━━━ 2.0s - 4.0s Shibuya Crossing. The nerdy tourist appears between several Japanese women. Everyone waves at camera. Selfie-stick perspective. GIRLS: 「イエーイ!」 TOURIST: 「よろしくお願いします!」 Fast cuts. ━━━━━━━━━━━━━━━━━━━━ 4.0s - 6.0s Cosplay district. Cosplayers surround the group. Peace signs. Camera spinning around them. GIRLS: 「写真撮ろう!」 TOURIST: 「もちろん!」 Rapid flash-frame transitions. ━━━━━━━━━━━━━━━━━━━━ 6.0s - 8.0s Arcade sequence. Button mashing. Bright arcade lights. Excited reactions. GIRL: 「すごい!」 TOURIST: 「やったー!」 Everyone cheers. Quick zooms. ━━━━━━━━━━━━━━━━━━━━ 8.0s - 10.0s Karaoke room. The tourist sings enthusiastically. Everyone claps and laughs. GIRLS: 「頑張れー!」 TOURIST: 「ありがとう!」 Japanese rock music blends with karaoke atmosphere. Fast jump cuts. ━━━━━━━━━━━━━━━━━━━━ 10.0s - 12.0s Late-night food alley. Group eating together. Street food. Laughing. GIRL: 「おいしい?」 TOURIST: 「めちゃくちゃ美味しい!」 Everyone laughs. ━━━━━━━━━━━━━━━━━━━━ 12.0s - 15.0s Final montage. Tokyo skyline. Neon streets. Group selfies. Walking together through nightlife district. Everyone talking over each other excitedly. GROUP: 「乾杯!」 Camera rushes forward through the crowd. Music hits the chorus. Abrupt energetic cut. END.

Seedance 2.0 AI video prompt: Use the uploaded images as the primary visual references.

Image role:

Image 1,

Use the uploaded images as the primary visual references. Image role: Image 1, Image 2, Image 3, Image 4, and Image 5 are all character references. Randomly rotate the visible character among all five references throughout the clip. Do not rely only on Image 1. Do not merge all characters into one design. Keep each character visually recognizable according to its own reference. Across the 15-second clip, all five reference characters should appear repeatedly in a fast, random-feeling order. 15-second music-video style animation. Ultra cute hyper pop, visually explosive, chaotic, surreal kawaii motion collage. No lip-sync. No singing focus. Prioritize visual impact, diverse chaos, lens effects, rapid switching, and striking composition changes. Main direction: Create a dense, hyperactive, kawaii-chaotic visual sequence where the characters from Image 1–5 appear in rapid rotation inside an aggressively stylized pop-surreal world. The attraction is visual impact: fisheye distortion, lens warping, smash zooms, split-panel chaos, clone echoes, kaleidoscope effects, sticker explosions, sudden camera angle changes, and cute chaotic dance/reaction bursts. The clip should feel playful, overwhelming, stylish, and highly memorable. Highest priority: Visual impact. Varied chaos. Strong lens effects. Rapid switching. Character rotation across Image 1–5. Cute but intense energy. Readable silhouettes and faces. Motion direction: Use cute but wild movement: short dance bursts, sudden pose changes, hopping, spinning, flailing arms in a playful way, head tilts, bouncing, crowd-like rhythm, exaggerated reactions, and spontaneous mob-dance energy. No lip-sync focus. Motion should feel energetic, slightly unhinged, rhythmic, and visually surprising. Chaos direction: Use many different kinds of chaos within one clip: fisheye lens distortion, ultra-wide lens warping, crash zooms, whip pans, dutch angles, rapid perspective shifts, kaleidoscopic mirroring, multi-panel fragmentation, repeated clones, echo trails, layered collage composition, sticker explosions, frame-within-frame effects, sudden close-up distortion, stretch-and-squash motion, graphic shock cuts, pop-art visual overload. Each few beats should feel visually different. Lens / camera effects: Strong fisheye close-ups. Wide-angle perspective exaggeration. Sudden zoom-ins and zoom-outs. Tilted framing. Quick rotations. Orbiting camera feel. Tunnel-like lens distortion. Occasional extreme close-up facial distortion in a cute, playful way. Use lens effects for impact, not realism. Visual style: Pastel neon palette, pink, cyan, lemon yellow, mint, lavender, plus occasional bold accent bursts. Bold outlines, flat shading, halftone dots, sticker symbols, flowers, bows, spirals, stars, geometric pop icons, abstract blobs, manga-style panels, candy-glitch accents, surreal layered collage. Cute chaos, not horror, not grotesque. Character usage: Rapidly alternate featured characters from Image 1–5. Some cuts may show one character in a fisheye close-up. Some cuts may show multiple characters in different panels. Some cuts may duplicate the same character several times in different positions. The sequence should feel like a rotating cast in a chaotic visual relay. Expression direction: Use strong expression variety: shock, joy, mischief, hyped excitement, wide-eyed overload, wink, cheeky grin, spinning dizzy face, playful smugness, surprised cute reaction. Expressions should feel punchy and stylized, but not tied to lip-sync. Editing: Very fast cuts, every 0.1 to 0.5 seconds. Use smash cuts, panel swaps, strobe-like graphic switches, sudden lens changes, snap zooms, repeated frame echoes, multi-panel face bursts, and visual rhythm that constantly mutates. Avoid long stable shots. Keep the sequence dense and surprising. Timeline: 0.0–2.0s Immediate visual explosion. A randomly selected character appears in an extreme fisheye or ultra-wide close-up. Pastel symbols, stickers, spirals, and graphic shapes burst into frame. Quick cut to another character with a different lens distortion. 2.0–4.0s Increase variety. Rapid rotation among Image 1–5. Use dutch angles, tunnel zoom, clone echoes, panel fragmentation, and pop-art overlays. Characters do brief dance-like or reaction-based motion bursts. Each cut should introduce a different type of visual chaos. 4.0–6.0s Split-panel chaos. The screen breaks into manga-style panels. Different characters appear in different panels at the same time. Some panels show face close-ups, some show small dance bursts, some show sticker explosions. Panels slide, rotate, and snap into new positions. 6.0–8.0s Mob-dance energy. Characters appear as repeated clones or quick alternating performers. Use synchronized bouncing, side-step gestures, tiny hops, hand poses, and playful crowd-like movement. The camera swings with a fisheye lens, making the dance feel warped and energetic. 8.0–10.0s Kaleidoscope section. Faces, stickers, bows, flowers, and abstract shapes mirror into a rotating kaleidoscope. Characters appear in mirrored fragments but remain cute and recognizable. The frame feels dense, colorful, and hypnotic. 10.0–12.0s Crash zoom reaction montage. Fast zoom-ins on eyes, faces, hands, and cute poses. Switch expressions rapidly: wide-eyed surprise, mischievous grin, wink, smug face, dizzy spiral-eye mood, and overexcited joy. Use sharp pop-art impact frames and halftone bursts. 12.0–13.5s Lens tunnel and graphic overload. The camera rushes through a tunnel of stickers, flowers, spirals, and pastel panels. Multiple characters pop in one after another using different lens effects: fisheye, wide-angle stretch, diagonal crop, rotating close-up, clone echo. The screen becomes a layered kawaii collage. 13.5–15.0s Final impact climax. Fastest switching, strongest lens effects, biggest sticker explosions, and the densest visual collage. Show several characters in rapid succession or in simultaneous panels. End on a strong high-impact final frame: distorted cute close-up, wild pose, or layered group-chaos composition. Important constraints: No lip-sync focus. No slow shots. No realistic concert staging. Do not let only one character dominate. Use many different visual tricks across the clip rather than repeating one effect. Keep the chaos playful, cute, and stylish rather than dark or grotesque. Negative prompt: realistic style, horror, gore, grotesque deformation, broken anatomy, extra limbs, muddy colors, low energy, empty background, plain static camera, weak motion, repetitive single-shot composition, one-character-only repetition, realistic performance stage

Seedance 2.0 AI video prompt: SINGLE SHOT, no multicut.

Use the video as a reference for the camera work, the

SINGLE SHOT, no multicut. Use the video as a reference for the camera work, the push-in and the general blocking of the elements in the scene. The image1 corresponds to the first frame and must serve as a visual guide for the initial composition of the shot. The red objects that rotate in the video represent asteroids. Use their movement, spin and flotation as a reference to animate the asteroids in space. The green square represents only the approximate position, scale in shot and spatial location of the astronaut regarding the camera. Use it only as a placement marker inside the framing. Completely replace the green square with an astronaut of image1 with helmet and spacesuit. The square must not appear in the final result. It must serve solely as a reference of where to place the astronaut in the scene. The scene shows asteroids floating in space while the camera does a push-in towards an astronaut with helmet and spacesuit. The astronaut floats in zero gravity and occupies the position marked by the green square, with a very subtle, slow and controlled body animation. The astronaut does not perform effusive or agitated movements. Remains floating with very slow movements, almost suspended, maintaining a sensation of passive drift in space. The arms move only very slightly, with soft gestures as if they were displacing backwards by inertia or trying to stabilize with minimal effort. The body remains calm, with slow and natural micro-adjustments of balance in zero gravity. The legs remain completely still during the whole video. There is no leg movement whatsoever. All the body animation must concentrate solely on small, slow and subtle movements of the arms. The head also moves with a lot of containment. If there is nervousness, it must be felt only in a subtle way, through small slow gestures, not through abrupt movements. The general sensation must be of an astronaut lost and floating adrift, but expressed with a minimal, slow and credible performance. Very important: maintain the same relative position in frame, scale in shot and relationship with the camera indicated by the green square, but substitute that marker with the complete astronaut. The green square is only a location guide and must disappear completely in the final render. Important: backlit illumination, not frontlit. An intense light from the background illuminates the scene, creating silhouettes, strong rim light on the asteroids and the astronaut, dramatic spatial glow and cinematic contrast. Photorealistic style, cinematic, with a background of stars, spatial depth and credible flotation movement.

Seedance 2.0 AI video prompt: Visual Style: Soft watercolor illustration, warm pastel palette, gentle painted

Visual Style: Soft watercolor illustration, warm pastel palette, gentle painted textures, thick storybook outlines, Pixar-meets-picture-book warmth. Every frame should look like it could be lifted from a printed children's book page. 0–3s: Storybook illustrated wide shot of a tiny golden town at sunrise. Cobblestone streets, pastel shop fronts, flowers in window boxes. Soft watercolor sky in peach and lavender. A warm narrator title card fades in like a book page turning. Camera gently pushes forward. No characters yet — just the world breathing awake. Painterly, soft, magical. Children's picture book aesthetic. 3–6s: Illustrated medium shot. Two heroic girls stroll into frame side by side down the golden street. The blonde girl has long wavy golden hair catching the morning light, bright blue eyes, wearing a blue superhero shirt with a red-and-gold crest, a flowing tan coat and red skirt. Beside her, a tall dark-haired girl in a full navy blue superhero suit with a flowing red cape. Both hold iced drinks with green straws. They walk in easy, comfortable silence. Warm amber light. Storybook illustration style — soft outlines, gentle cel-shading, children's book warmth. 6–9s: Storybook close-up at child's eye level. A tiny girl with pigtails presses her nose against a bakery window, eyes wide with wonder as the two superhero cousins pass outside. Illustrated golden reflections on the glass. Warm, tender, magical. The moment feels frozen — like a painting inside a book. No dialogue. Just wonder. 9–12s: Illustrated wide shot. The warm golden sky above the town slowly shifts. A single large storybook storm cloud — illustrated with soft curling edges, not scary, like a cloud from a picture book — drifts in from the horizon. The cousins look up slowly. Their expressions go calm and ready. Camera tilts upward toward the cloud. The warm light dims slightly to cool lilac. Still painterly and safe — this is a children's story. 12–15s: Slow illustrated close-up. Two iced drinks placed gently side by side on a low stone wall. Camera pulls back in slow motion. The cousins stand tall, side by side, capes beginning to lift in the rising wind. Their silhouettes are bold and warm against the shifting sky. The frame freezes — like the last illustration before a chapter turn. Storybook page texture overlays softly. Painterly. Heroic. Tender.

Seedance 2.0 AI video prompt: 빠른 컷 전환 템포, 핸드헬드 카메라 무브먼트

@캐릭터 의 얼굴 모든 구간에서 완전히 일치시킬 것.
착장, 헤어스타일은 구간마다 완전히 다른

빠른 컷 전환 템포, 핸드헬드 카메라 무브먼트 @캐릭터 의 얼굴 모든 구간에서 완전히 일치시킬 것. 착장, 헤어스타일은 구간마다 완전히 다른 스타일로 교체. 0-2초: 【침실 거울】 크롭 화이트 티 + 와이드 데님. 거울 셀카 풀샷. 카메라로 빠르게 줌인. 2-4초: 【카페 실내】 플로럴 미디 드레스 + 메리제인. 라떼 집는 손 클로즈업 → 얼굴 리액션 빠른 컷. 4-6초: 【쇼핑몰 피팅룸】 Y2K 세트업. 옷 들고 포즈. 정면·측면 2컷 빠르게 전환. 6-8초: 【편의점】 오버사이즈 후드 + 카고팬츠. 냉장고 앞 손 클로즈업 → 음료 마시는 입 클로즈업. 8-10초: 【도심 거리】 블랙 미니스커트 + 버클 부츠. 핸드헬드 팔로우샷. 발걸음 → 얼굴 빠른 줌인. 10-12초: 【루프탑·옥상】 새틴 슬립 드레스 + 레이어드 체인 네크리스. 바람에 드레스 휘날림. 골든아워 역광. 12-15초: 【야외 골목】 케이팝 아이돌 룩 — 크롭 재킷 + 미니 스커트 + 선글라스. 정면 풀샷에서 카메라 향해 천천히 걸어옴. 마지막 프레임 정지. 카메라: 0.5~1초 빠른 컷 리듬. 핸드헬드 아이폰 질감. 줌인·줌아웃·틸트업 혼용. 자연스러운 흔들림. 사운드: 경쾌한 시티팝 BGM. 카페 소음·발소리·옷 스치는 소리 환경음 레이어. 스타일: 밝고 생기있는 자연광 톤. 필름 룩 약하게. 인스타그램 브이로그 감성. 임의로 컷을 통합하거나 장면을 생략하지 않는다.

Seedance 2.0 AI video prompt: Scene:
A seamless ultra-cinematic one-take shot starting from deep space and end

Scene: A seamless ultra-cinematic one-take shot starting from deep space and ending in an intimate café moment. The sequence emphasizes speed, scale, and immersive transition from cosmic to human scale. Subject / Character: Final subject is — a young woman sitting in an open-air cafe named "AinaAiTech", wearing denim shorts and a white shirt, casually eating a hamburger with a drink beside her. Action Timeline (TOTAL: 15s): 0–4s (Space → Earth Approach): Wide cinematic shot of Earth in deep space. The camera accelerates forward smoothly. Earth rapidly grows in frame. Subtle light streaks, atmospheric glow becomes visible. Motion builds gradually but feels grand. 4–8s (Atmospheric Entry → Fast Descent): Camera pierces atmosphere with intense glow and motion blur. Clouds rush past. Continents and terrain sweep underneath at high speed. Strong sense of acceleration. 8–11s (City Dive → Street Flow): The camera locks onto a city and dives sharply. Skyscrapers rise fast. Transition into street-level movement — fluid glide through streets, passing buildings, corners, and urban elements with dynamic motion. 11–12s (Café Target Lock): The camera spots an outdoor cafe with a large "AinaAiTech" sign. Rapid but smooth deceleration begins. Focus tightens. 12–15s (Final Scene → Character): The camera settles into a medium shot of the character. She sits casually, eating a hamburger, with a drink on the table. Natural motion (taking a bite, relaxed posture). Warm, calm contrast to the previous high-speed sequence. Camera: One continuous shot, no cuts. Extreme speed ramping: slow → ultra-fast → controlled slowdown. Wide lens in space → natural cinematic lens at final shot. Smooth stabilization with slight handheld realism at the end. Audio: Cinematic rise from deep space ambience → intense whooshing during descent → city ambience → soft café sounds (light chatter, ambient noise). Style: Cinematic color grading: cool tones in space/descent → gradually warmer tones at street/café. High contrast, subtle film grain, volumetric lighting, atmospheric particles, ultra-detailed, 8K, photo-realistic.

Seedance 2.0 AI video prompt: Use the provided (@ Reference Image) character reference as the main subject for

Use the provided (@ Reference Image) character reference as the main subject for Chris First. Keep his face, hairstyle, body type, and the exact same outfit from the reference. Do not redesign his clothing, colors, shoes, accessories, or overall look. Create a 15-second realistic televised wrestling match moment in the style of a live Monday Night Raw broadcast. The entire video takes place inside the wrestling ring. Do not start with a high-angle arena shot, entrance shot, or cinematic establishing shot. The camera should already be focused on the in-ring action when the video begins. The shot should feel like a single continuous TV broadcast take, filmed from a realistic ringside hard-camera angle. The camera can subtly pan, zoom, and adjust focus like a real live wrestling broadcast, but it should not cut to dramatic movie angles. Keep both wrestlers visible and make the action clear, grounded, and believable. Chris First is facing The Undertaker in the ring. The Undertaker is tall, imposing, and dressed in dark wrestling gear, with a serious, intimidating presence. Chris First is intense, focused, and athletic. The action starts immediately, as if the match has already been going on. The crowd is visible behind the ring ropes throughout the shot. Fans are standing, cheering, and holding signs supporting Chris First. Several signs should clearly read “WE LOVE FIRST”, while others say “FIRST IS THE FUTURE”, “CHRIS FIRST!”, and “FIRST TO WIN”. The signs should look like real handmade wrestling crowd signs, not perfect digital graphics. Chris First and The Undertaker are already locked up near the center of the ring. Chris First slips behind The Undertaker, fights for control, and uses momentum to hit a basic, realistic wrestling body slam. The move should look like a standard professional wrestling move, not an RKO, not a finishing move, and not overly stylized. Make it clean, physical, and believable for live TV. Timeline 0:00–0:03 The video begins already inside the ring from a realistic TV hard-camera angle. Chris First and The Undertaker are grappling immediately. The crowd is visible behind the ropes, with handmade signs reading “WE LOVE FIRST” and “CHRIS FIRST!” in the background. 0:03–0:06 Chris First pushes back against The Undertaker, ducks under one arm, and shifts position to gain leverage. The camera subtly follows the movement like a live broadcast operator tracking the action. The Undertaker looks powerful, but Chris First is quicker. 0:06–0:10 Chris First gets control, hooks The Undertaker, and executes a realistic basic body slam in the center of the ring. The ring mat reacts with a believable bounce on impact. The crowd rises and cheers as the move lands. 0:10–0:13 The Undertaker is down on the mat. Chris First takes one step back, breathing hard, focused and confident. The camera slightly zooms in the way a live wrestling broadcast would after a big momentum shift. 0:13–0:15 Chris First raises one arm and looks out toward the crowd. Fans behind him are cheering and waving signs that say “WE LOVE FIRST” and “FIRST IS THE FUTURE”. End with Chris First standing tall in the ring, still in the same continuous broadcast-style shot. Visual Style Realistic Monday Night Raw-style live TV broadcast, continuous in-ring camera coverage, no cinematic cuts, no high-angle opening, no entrance sequence, bright arena lighting, real crowd energy, visible handmade fan signs, authentic wrestling ring physics, grounded professional wrestling movement, natural camera zoom and pan, live sports entertainment presentation. Narrator Voice Dialogue — Separate From Timeline “Chris First is taking the fight right to The Undertaker! Look at the quickness, look at the control — Chris First turns it around! He’s got him up… and slams him down in the middle of the ring! The crowd is on its feet for Chris First!”

Seedance 2.0 AI video prompt: Use the uploaded images as character and style references.

15-second kawaii hyp

Use the uploaded images as character and style references. 15-second kawaii hyper-pop music video at BPM 174. No lip-sync. No singing focus. Create a fast, chaotic, cute visual sequence with rapid character switching among all uploaded reference characters. Style: Pastel neon anime, pink, cyan, yellow, mint, lavender, bold outlines, flat shading, stickers, flowers, bows, spirals, halftone dots, manga panels, candy glitch accents. Main motion: Fast BPM174 rhythm, rapid cuts every 0.1–0.4 seconds, cute dance bursts, sudden poses, bouncing, spinning, mob-dance energy, exaggerated reactions, clone echoes, split panels, and visual overload. Lens effects: Use strong fisheye and ultra-wide distortion. Make the center bulge forward. Stretch and curve the edges clearly. Use obvious barrel distortion, warped perspective, tunnel zooms, crash zooms, rotating camera, dutch angles, whip pans, and sudden push-ins. Variation: Do not repeat similar shots. Each section should introduce a different effect: fisheye close-up, spinning camera, split-panel shatter, clone multiplication, kaleidoscope mirror, mob-dance burst, sticker tunnel, wide-angle distortion, radial explosion, multi-character chaos. Timeline: 0–3s: extreme fisheye close-ups, fast character switching, sticker bursts. 3–6s: rotating camera, panel splits, shattered manga frames. 6–9s: clone multiplication, mob-dance bouncing, wide-lens warping. 9–12s: kaleidoscope mirrors, tunnel zooms, crash zoom reactions. 12–15s: maximum chaos, rapid effect relay, all characters in fast succession, final distorted cute impact frame. Important: Make the fisheye distortion visually obvious, especially at the edges. Keep faces cute and readable. Use many different visual tricks, not one repeated effect. Negative prompt: weak fisheye, only round frame, static camera, repetitive shots, slow motion, lip-sync, realistic style, horror, grotesque face, broken anatomy, extra limbs, muddy colors, low energy

Seedance 2.0 AI video prompt: Create a cinematic 15-second anime-style video inside a huge abandoned industria

Create a cinematic 15-second anime-style video inside a huge abandoned industrial warehouse with tall windows, metal beams, skylights, dusty air, dramatic sunbeams, and wet reflective puddles on the concrete floor. The video opens with a very wide establishing shot of the warehouse that lasts only 1 second. During that first second, the camera sees the glowing arcade football machine far in the distance, surrounded by a group of France national-team-inspired players and close friends. Right after that first second, the camera immediately pushes in with a fast zoom/dolly-in toward the group, creating a sudden rise in energy and focus. Do not use real names, real logos, club logos, federation logos, or official branding. The main character is an intense anime-style adult French forward inspired by a superstar number 10: athletic build, short cropped dark hair, sharp focused eyes, dark blue national-team-inspired jersey, white shorts, red socks, and fast confident energy. He leans at the arcade machine, fully locked into the game. His teammates and close friends stand tightly around him in matching dark blue kits, cheering, shouting, holding their breath, and reacting like a squad supporting him in the final seconds of a match. The arcade screen clearly shows a fictional football video game match: FRANCE vs PARAGUAY, score 0–0, timer near the final minute, mini-map, shot power bar, player indicator for number 10, and fast attacking gameplay. France plays in dark blue, Paraguay plays in red-and-white striped kits. The main character grips the joystick and presses the buttons with intense speed, his body leaning forward with pressure, excitement, and belief, as if he is trying to score the winning goal. Camera flow: 0–1 second is a very wide warehouse establishing shot. At exactly 1 second, the camera immediately rushes forward with a fast zoom/dolly toward the arcade machine and the surrounding players. Then cut into a side-angle medium shot of the main character at the controls, surrounded by teammates. Show an extreme close-up of his hands on the joystick and buttons, matching the same hand position from the side shot: left hand firmly gripping the joystick from the side, right hand hovering over and hammering the arcade buttons, forearms tense, blue arcade glow reflecting on the skin and controls. Then push into a frontal close-up of his determined face, sweat on his brow, eyes reflecting the game screen. Show teammates behind him shouting and encouraging him. Move to an over-the-shoulder shot where the arcade screen shows the final France attack against Paraguay. Then show the teammates reacting with rising tension, fists clenched and mouths open. Finish with the arcade screen showing number 10 taking the decisive shot toward goal, the power bar maxed, goalkeeper diving, then a final hero shot of the main character still at the machine while everyone explodes with celebration around him. Audio/dialogue: energetic arcade sounds, fast button clicks, joystick movement, crowd noise from the game screen, teammates shouting in the background. The main character quietly hypes himself up in Japanese with short emotional spoken lines, no subtitles: “いける…まだいける!” “集中しろ…” “俺ならできる!” “よし、今だ!” “決める!” The tone should feel like a dramatic sports anime moment, full of tension, friendship, and arcade excitement. Visual style: high-quality cinematic anime, expressive faces, dynamic camera movement, dramatic blue arcade glow mixed with warm warehouse sunlight, shallow depth of field, motion blur on fast button presses, clean detailed linework, emotional sports-anime energy, no photorealism, no text overlays, no subtitles, no real brand logos, no official emblems, no real player names.

Seedance 2.0 AI video prompt: Use the uploaded images as the primary visual references for character design, f

Use the uploaded images as the primary visual references for character design, facial style, color palette, graphic motifs, and overall kawaii-chaotic tone. 15-second music-video style animation, ultra cute and hyper pop, fast-paced, candy-colored, chaotic but readable, anime-inspired graphic motion. Main direction: The characters dance in a cute, playful, slightly chaotic “mob dance” / “idol crowd dance” feeling. The motion should feel inferred naturally from the reference images, not rigid choreography. Emphasize fast switching expressions, group-like energy, rapid cut transitions, and a dense pop visual atmosphere. Visual style: Pastel neon palette, pink, cyan, lemon yellow, mint, lavender. Bold clean outlines, flat shading, sticker-like icons, bows, flowers, sparkles, spirals, abstract shapes, pop symbols, glitchy accents, halftone dots, split panels, manga-like framing, layered collage feeling. Cute and stylish chaos, not horror, not grotesque. Character motion: Generate adorable dance movement with small bouncy steps, side-to-side body sway, quick hand poses, synchronized hops, idol-style gestures, playful pointing, little spins, shoulder pops, head tilts, wink poses, cheerful bouncing, cute rhythmic stepping. Include moments that feel like a crowd dance or mob dance, even if only one or a few characters are visible at a time. The dance should feel energetic, spontaneous, and rhythm-driven. Expression direction: Rapid expression changes: smile, wink, surprised face, excited open-mouth face, mischievous grin, hypnotic stare, sparkling eyes, dizzy spiral-eye mood, overjoyed face, playful chaos. Expressions should switch quickly with the cuts, giving a hyperactive kawaii feeling. Editing / camera: Very fast-cut visual rhythm. Use rapid cut-ins, jump cuts, smash zooms, push-ins, pull-backs, diagonal framing, split-screen panels, close-ups of faces, occasional medium shots for dance, and layered pop composition. Cuts can change every 0.2 to 0.8 seconds. Alternate between close-up face shots, upper-body dancing, and brief group/crowd-energy compositions. The camera should feel lively and music-synced. Scene flow: 0–3s: Introduce the character(s) with fast close-ups, blinking, cute smile, eye highlights, bows/flowers/icons popping in, fast panel cuts. 3–6s: Start the dance with small synchronized idol-like motions, side steps, hand poses, bouncing shoulders, quick expression switches. 6–9s: Increase the chaos. More split panels, more characters or crowd-energy feeling, fast alternating expressions, quick hops, spins, and rhythmic pose changes. 9–12s: Peak energy. Cute mob-dance vibe, dense visual layering, multiple rapid cut-ins, close-up reaction faces, synchronized bounce and gesture-heavy movement. 12–15s: Final climax. Maximum kawaii chaos with energetic dancing, smiling faces, wink, big pose finish, layered pop symbols exploding around the frame, ending on a strong cute final pose. Important priorities: Keep the characters cute and appealing at all times. Preserve readability of faces and silhouette. Let the dance and expressions feel naturally inferred from the references. The chaos should come from editing, framing, layered motifs, and rapid expression changes, not from broken anatomy or visual confusion. Negative prompt: horror, dark tone, realistic style, grotesque face, body deformation, extra limbs, broken hands, scary expressions, dull colors, muddy palette, slow motion, empty background, overly realistic physics, unreadable face, low energy, depressing mood.Make the choreography feel partially AI-inferred from the visual mood of the references: cute, chaotic, rhythmic, slightly unpredictable, but always charming and musically coherent.

Seedance 2.0 AI video prompt: A single uninterrupted cinematic shot begins in absolute darkness. There is no l

A single uninterrupted cinematic shot begins in absolute darkness. There is no light, no stars, no sound except an almost inaudible deep cosmic rumble. The camera slowly glides forward through the infinite black void, creating a feeling of mystery and anticipation. At exactly the third second, the first spark of light appears in the distance. It rapidly expands into an enormous stellar explosion, releasing billions of glowing particles that scatter across the universe. The particles become newborn stars, each igniting one after another like a chain reaction until an endless field of stars fills every direction. Massive colorful nebulae bloom from clouds of cosmic dust in brilliant shades of deep blue, emerald, crimson, violet, and gold. Without a cut, the camera accelerates smoothly between the stars. Entire galaxies spiral into existence around the viewer. Their luminous arms rotate naturally with realistic gravitational motion while glowing gas clouds stretch through space. Black holes bend light with accurate gravitational lensing, warping distant galaxies as the camera passes dangerously close before escaping their pull. The camera now races toward one bright spiral galaxy. It dives directly through the glowing galactic core without collision, emerging into a young solar system where countless molten planets orbit a blazing newborn sun. Streams of glowing asteroids and shimmering comets cross the frame naturally, adding scale and depth. The camera locks onto one fiery planet. As it approaches at incredible speed, the molten surface grows larger until it fills the frame. Volcanic eruptions throw rivers of lava into the atmosphere. Gigantic tectonic plates shift violently while oceans of magma ripple realistically beneath thick volcanic smoke. Without cutting, time accelerates dramatically. The molten planet cools before the viewer's eyes. Lava hardens into dark volcanic rock. Steam erupts across the entire surface as billions of years pass in seconds. Torrential rain begins, falling continuously until vast oceans rapidly form. Clouds swirl around the atmosphere while continents slowly rise from beneath the water through realistic plate tectonics. Mountain ranges thrust upward. Rivers carve valleys. Forests spread like green veins across the land. Polar ice caps form. The once-hostile world transforms into a vibrant blue planet full of life. The camera continues descending through the atmosphere in one perfectly smooth movement. It flies over towering mountain peaks, ancient forests covered in morning mist, winding rivers reflecting golden sunlight, vast oceans with rolling waves, and clouds illuminated by soft sunrise light. At the climax, the camera dives toward a peaceful grassy field where a young child stands looking upward. The child slowly raises their head toward the sky as sunlight illuminates their face. The camera gently pushes into one eye until the iris fills the entire frame. Inside the reflection of the eye, the entire universe is visible—galaxies, nebulae, stars, and planets—revealing that the cosmos lives within human curiosity. The final frame holds for one second as the pupil becomes a perfect circular portal containing the infinite universe before fading softly to black. CAMERA One continuous unbroken shot. No cuts. No teleportation. Smooth IMAX-style drone movement. Natural acceleration and deceleration. Extreme scale transitions. Perfect stabilization. Wide-angle cinematic lens gradually transitioning into macro close-up. Elegant motion with physically believable inertia. LIGHTING Physically accurate cosmic lighting. Volumetric god rays. High dynamic range. Glowing nebula illumination. Molten lava emission. Golden sunrise atmosphere. Realistic atmospheric scattering. Soft cinematic bloom. Deep shadows with rich contrast. ATMOSPHERE Epic. Awe-inspiring. Emotional. Scientific realism blended with cinematic wonder. A sense of witnessing creation itself. COLOR PALETTE Deep cosmic blacks. Electric blues. Fiery oranges. Molten reds. Emerald greens. Golden sunrise. Natural Earth tones. Rich HDR contrast. QUALITY Ultra photorealistic. True cinematic physics. 8K. IMAX documentary quality. Hyper-detailed textures. Volumetric particles. Realistic fluid simulation. Physically accurate clouds. Ray-traced reflections. Filmic depth of field. Natural motion blur. Premium visual effects. No CGI appearance. NEGATIVE PROMPT No text. No logos. No subtitles. No low-resolution textures. No flickering. No frame interpolation artifacts. No cartoon style. No unrealistic camera shakes. No duplicated objects. No broken planetary geometry. No oversaturated colors. Maintain perfect temporal consistency throughout the continuous shot.

Seedance 2.0 AI video prompt: SCENE CONTEXT
An elegant young woman films herself on her phone across one eveni

SCENE CONTEXT An elegant young woman films herself on her phone across one evening out: a short front-camera selfie moment in a luxury hotel room, a mirror selfie in an elevator, a walk past a boutique window at dusk, and the longest part — a dim candlelit restaurant where she films her reflection in a wall mirror across the table. Casual self-admiring video diary, calm and confident. ACTIVE REFERENCES @image1: young woman, mid-20s, long brown wavy hair, black strapless corset dress with subtle textured shimmer, diamond tennis necklace, butterfly diamond stud earrings, soft glam makeup. 100% matches the reference. Same dress, hair and jewelry in every cut. FORMAT MODE Timed multishot, four segments, HARD CUT between each. Inside each segment: one continuous take, single steady framing, camera position held from the segment's first frame to its last. 0.0–2.5s — CUT 1 hotel room, front camera 2.5–5.5s — CUT 2 elevator mirror 5.5–9.0s — CUT 3 evening street, boutique window reflection 9.0–15.0s — CUT 4 restaurant, wall mirror across the table OPTICS All segments: 84° wide FOV, clean modern phone lens, deep focus, everything sharp. The lens renders the frame evenly — brightness and shadow come only from the scene's own lighting. No drift mid-segment. CAMERA Every cut is footage from her own phone held in her hand. Her arm is steady and relaxed: only a soft natural sway under 1 cm, smooth and continuous, the frame floats gently at one fixed framing per segment. CUT 2: mid-segment she does one quick digital punch zoom-in on her face, then back out. CUT 4: slow digital zoom-in toward her reflection over the last 2 seconds. ACTION CUT 1 — a brief selfie beat, phone out of frame, front camera at arm's length slightly above eye level: she stands in one spot in the hotel room, crystal chandelier above, holds her pose calmly, tilts her chin once, light closed-lip smile. CUT 2 — elevator interior, warm metallic walls, she stands facing the elevator mirror holding the phone horizontally in both hands at chest height, phone visible in the reflection; she shifts weight to one hip, adjusts the necklace with two fingers, calm gaze into the mirror. CUT 3 — evening street at dusk: she walks at 4 km/h parallel to a large lit boutique window on her left. She holds the phone at chest height in her right hand, screen toward her, lens pointed at the glass at a slight 45° angle ahead of her, and keeps this exact grip and angle for the whole segment. The frame shows the shop window: her sharp reflection walks through it alongside lit mannequins, the phone visible in her hand in the reflection, warm shop light behind the glass, her hair moving with her steps. CUT 4 — the longest, calmest part. Dim restaurant, low warm light. She sits at a table FACING a large wall mirror mounted 2.5 meters IN FRONT of her, on the far side of the table; her phone films that mirror straight ahead. The frame shows the mirror with her seated reflection at medium distance, phone in her hands visible in the reflection; a lit candle and a water glass on the table in the foreground. She has time here: straightens her posture, tucks a strand of hair behind her ear, rests her chin lightly on her free hand, relaxed half-smile at her own reflection. Final frame: her calm smiling reflection in the candlelight, sharp and centered — natural thumbnail. PERFORMANCE Restrained, self-assured, unhurried: micro-movements only, soft closed-lip smile, relaxed brows, living eyes with catch-lights from chandelier / elevator lamps / street lights / candle flame. Pore-level skin realism, natural skin texture, fine flyaway hairs catching light. PHYSICS Hair sways with real inertia on every head turn and step. Dress fabric shifts naturally as she moves and sits. Necklace and earrings catch and scatter light with her motion. Reflections in elevator mirror, shop glass and restaurant mirror perfectly synced to her movement. Candle flame flickers gently, its warm light dancing on her jewelry. LIGHTING CUT 1: warm interior key from the chandelier above, 3800K, soft bounce from cream walls. CUT 2: even warm elevator downlight, 3500K, gentle specular streaks on metallic walls. CUT 3: dusk ambient 5600K sky mixed with warm 3200K shop-window glow spilling onto her face. CUT 4: low-key dim dining room — the candle is the key light, 2800K warm flame glowing up onto her face, small pools of warm practical lamps deep in the background, the rest of the room falls into soft dark shadow; intimate hushed evening mood, her face and jewelry the brightest points of the frame against the dark interior. AUDIO Live natural sound only, changing per cut: quiet hotel room tone; low elevator hum and a soft ding; evening street ambience, her heels on pavement, distant traffic; then quiet restaurant — soft murmur of guests, occasional cutlery clink, low mellow music bed. STYLE Authentic modern phone-camera footage: photoreal, high-detail source, natural color response, true skin tones, very fine subtle grain only, real-time speed in every segment, 16:9 horizontal. OUTPUT SETTINGS 16:9 horizontal, real-time speed in all four segments. POSITIVE LOCKS Same woman from @image1 in all four cuts: identical face, hair, black corset dress, diamond necklace and butterfly earrings. Phone visible in her hand in every mirror/window reflection, phone out of frame in CUT 1. Each segment is one continuous take with one fixed framing — steady relaxed handheld float under 1 cm from first frame to last. CUT 1 lasts 2.5 seconds only. CUT 4 is the longest segment (6 seconds), dim candlelit mood, her face lit by the candle as the brightest point of the frame, wall mirror across the table 2.5 meters in front of her, her reflection at medium distance. Calm light smile as the only emotion. Evening timeline consistent: warm interior → elevator → dusk street → dim candlelit restaurant.

Seedance 2.0 AI video prompt: Use the uploaded images as the primary visual references for character design, f

Use the uploaded images as the primary visual references for character design, facial style, color palette, graphic motifs, and overall kawaii-chaotic tone. 15-second music-video style animation, ultra cute and hyper pop, fast-paced, candy-colored, chaotic but readable, anime-inspired graphic motion. Main direction: The characters dance in a cute, playful, slightly chaotic “mob dance” / “idol crowd dance” feeling. The motion should feel inferred naturally from the reference images, not rigid choreography. Emphasize fast switching expressions, group-like energy, rapid cut transitions, and a dense pop visual atmosphere. Visual style: Pastel neon palette, pink, cyan, lemon yellow, mint, lavender. Bold clean outlines, flat shading, sticker-like icons, bows, flowers, sparkles, spirals, abstract shapes, pop symbols, glitchy accents, halftone dots, split panels, manga-like framing, layered collage feeling. Cute and stylish chaos, not horror, not grotesque. Character motion: Generate adorable dance movement with small bouncy steps, side-to-side body sway, quick hand poses, synchronized hops, idol-style gestures, playful pointing, little spins, shoulder pops, head tilts, wink poses, cheerful bouncing, cute rhythmic stepping. Include moments that feel like a crowd dance or mob dance, even if only one or a few characters are visible at a time. The dance should feel energetic, spontaneous, and rhythm-driven. Expression direction: Rapid expression changes: smile, wink, surprised face, excited open-mouth face, mischievous grin, hypnotic stare, sparkling eyes, dizzy spiral-eye mood, overjoyed face, playful chaos. Expressions should switch quickly with the cuts, giving a hyperactive kawaii feeling. Editing / camera: Very fast-cut visual rhythm. Use rapid cut-ins, jump cuts, smash zooms, push-ins, pull-backs, diagonal framing, split-screen panels, close-ups of faces, occasional medium shots for dance, and layered pop composition. Cuts can change every 0.2 to 0.8 seconds. Alternate between close-up face shots, upper-body dancing, and brief group/crowd-energy compositions. The camera should feel lively and music-synced. Scene flow: 0–3s: Introduce the character(s) with fast close-ups, blinking, cute smile, eye highlights, bows/flowers/icons popping in, fast panel cuts. 3–6s: Start the dance with small synchronized idol-like motions, side steps, hand poses, bouncing shoulders, quick expression switches. 6–9s: Increase the chaos. More split panels, more characters or crowd-energy feeling, fast alternating expressions, quick hops, spins, and rhythmic pose changes. 9–12s: Peak energy. Cute mob-dance vibe, dense visual layering, multiple rapid cut-ins, close-up reaction faces, synchronized bounce and gesture-heavy movement. 12–15s: Final climax. Maximum kawaii chaos with energetic dancing, smiling faces, wink, big pose finish, layered pop symbols exploding around the frame, ending on a strong cute final pose. Important priorities: Keep the characters cute and appealing at all times. Preserve readability of faces and silhouette. Let the dance and expressions feel naturally inferred from the references. The chaos should come from editing, framing, layered motifs, and rapid expression changes, not from broken anatomy or visual confusion. Negative prompt: horror, dark tone, realistic style, grotesque face, body deformation, extra limbs, broken hands, scary expressions, dull colors, muddy palette, slow motion, empty background, overly realistic physics, unreadable face, low energy, depressing mood

Seedance 2.0 AI video prompt: 15-second absurd over-the-top Indian action movie scene, ultra-realistic, photor

15-second absurd over-the-top Indian action movie scene, ultra-realistic, photorealistic blockbuster footage, Telugu/Tamil mass cinema energy, ridiculous dramatic camera work, extreme zooms, impossible hero worship, epic soundtrack for a completely normal situation. 0:00 - 0:03 A young man enters a family gathering. Silence. He looks around the room. Slowly. Calmly. Then asks: "What's the Wi-Fi password?" Instantly, everyone's face freezes. SFX "DHOOOOOM!" "BWAAAAAAH!" Thunder strike. Glass rattles. 0:03 - 0:06 Extreme zoom into Grandma's eyes. Zoom into Uncle's eyes. Zoom into Dad's eyes. Zoom into the router. Zoom into the blinking internet light. Zoom back into Grandma's eyes. Each zoom gets a dramatic impact sound. SFX "SHING!" "TAK!" "BWAAAH!" 0:06 - 0:10 Grandma slowly stands up. The camera circles around her. Wind suddenly appears indoors. Curtains fly dramatically. Her saree moves like she's entering a battlefield. She whispers: "Only one man knows the password..." Massive bass drop. 0:10 - 0:13 Cut to Dad sitting silently in another room. The camera introduces him like the final boss. 50 rapid zooms. Slow-motion beard stroke. Lightning flashes outside. The router lights glow brighter. Music Epic Indian choir. Heavy drums. Electric guitar. 0:13 - 0:15 Dad slowly looks into camera. Adjusts his glasses. Whispers: "Password123" The entire house explodes into applause. Fireworks outside. Triumphant orchestral music. TITLE CARD "THE WIFI PASSWORD" COMING SOON Final SFX Lion roar + thunder + jet engine + crowd cheering.

More use cases

Frequently asked questions

What are the best AI video prompts for zoom shots?

The best zoom prompts name four things together: the zoom type (crash zoom, slow push-in, snap zoom, pull-back, dolly zoom), the direction and speed (toward the subject over three seconds, away from the subject over five seconds, a single-frame crash), the trigger beat (what moment in the scene causes the zoom to begin), and the starting and ending frames (medium establishing shot to extreme close-up, wide aerial to close on the subject's hands). Without all four, Seedance produces a generic focal-length shift without a dramatic purpose. Every prompt in this gallery encodes that four-part structure.

What's the difference between a zoom and a dolly shot in Seedance 2.0?

A zoom changes the lens focal length — the background scale changes relative to the subject, compressing toward them as you zoom in. A dolly moves the camera body through space — the spatial relationship between subject and background stays proportional, only the distance changes. These are not interchangeable: 'zoom in' and 'push-in / dolly in' produce different visual effects. The zoom effect is characterized by background compression; the dolly effect by the parallax between foreground and background elements. Name the one you want explicitly: 'zoom in' for the focal-length effect, 'dolly in' or 'push-in' for the physical camera move.

Can Seedance 2.0 generate the Vertigo / dolly zoom effect?

Yes. The Vertigo effect — also called the dolly zoom or Hitchcock zoom — requires naming both simultaneous movements explicitly: 'the camera dollies backward while simultaneously zooming in to maintain the subject's apparent size constant — the background scale shifts dramatically, the environment seeming to stretch or collapse behind the subject.' Without naming both the backward dolly and the compensating zoom-in together, Seedance executes only one movement and the Vertigo effect doesn't appear. The effect works best in a wide-angle environment where the background contains strong geometric lines (hallways, corridors, staircases, columns) that make the scale distortion visually legible. Browse the zoom prompts here for examples with preview videos.