Realistic AI Video Prompts

Hardware artifact specs, ban-list techniques, physical simulation priority, and live-event camera realism — these Seedance 2.0 realistic video prompts solve the most common failure mode in AI video generation: footage that looks unmistakably generated.

Realistic AI video fails for a structural reason, not a resolution reason. The model does not default to "looks fake" because it lacks training data on real footage — it defaults to "looks fake" because the prompt describes what should be in the frame instead of how the camera records it. "A woman walking through a market" is a subject instruction. The camera instruction — what rig, what mount, what lens, what stabilization, what exposure behavior — is missing entirely, and without it, Seedance fills the gap with its default output: impossibly stable, clean, color-graded footage that no real camera produces and that the viewer recognizes as AI the moment they see it. The camera register is the primary realism driver in AI video, not the subject. The same scene — a person in a busy street — reads as real footage or as synthetic based almost entirely on the camera: a wide handheld with autofocus breathing and slight focus hunting reads as a real human holding a camera; a perfectly stable, color-corrected wide shot reads as AI. The difference is not the content; it is the camera's physical behavior in the world. The most reliable technique for encoding real camera behavior is hardware specification. Naming the recording format activates its full artifact bundle simultaneously. Mini DV brings focus hunting, tape-compression soft grain, auto-exposure shifts, and consumer-lens color science. iPhone 14 Pro brings HDR processing, rolling shutter wobble on fast pans, autofocus breathing, and the 26mm-equivalent main camera's barrel distortion. GoPro Hero brings wide-angle fisheye, aggressive electronic image stabilization that slightly over-corrects, and the latitude behavior of small sensors in high-contrast light. Each of these is a complete camera grammar that a single spec word activates — which is why "mini DV" outperforms "handheld video with grain and slight shakiness" as a realism direction. The shorthand encodes the full physical system. Exclusion lists are the second core technique, and they are load-bearing in a way that positive descriptions are not. AI video generation defaults toward "improved" footage: stabilized, color-graded, multi-angle, subtitle-captioned, beauty-filtered, skin-smoothed. These improvements are baked into training data as signals of "good video" — so the model produces them unless directed otherwise. The exclusion list overrides the default set explicitly: "no cuts, no third-person shots, no cinematic color grading, no beauty filters, no skin smoothing, no artificial HDR, no on-screen text." Every item on that list is a specific default behavior that reads as AI-generated, and banning each one by name is more reliable than positively describing rawness, because "rawness" is an aesthetic judgment while "no color grading" is a constraint. Physical constraint as camera direction follows the same logic as hardware spec: instead of describing a style of camera movement, describe the physical geometry that makes the shot possible. "One hand holds the phone at arm's length, front-facing camera, the arm never leaves the frame" is a geometric constraint that limits what the camera can physically do — and the model, given that constraint, stops generating impossible coverage (alternate angles, drone cuts, stable shots from across the room). The constraint is more effective than a style instruction because it restricts the model's solution space rather than merely preferences it. Physical simulation requirements address the four material systems AI video fails on most visibly: granular surfaces (sand, dust, snow — individual particles that scatter in physically correct trajectories when disturbed), soft body (cloth and hair — fabric that lags behind motion by the correct amount for its weight, hair that follows body movement with secondary motion), fluid impact (splash and pour sequences — real splash physics are smaller and faster than AI defaults, which produce spectacular slow-motion water effects that read as synthetic), and fluid depth and clarity (water that is transparent to the correct depth, refracts light in the correct pattern, and carries the right surface specular). Naming all four systems explicitly tells Seedance which simulations to prioritize, and adding "no exaggeration" to fluid systems calibrates against the default toward spectacular effects. The real-video format recognition technique completes the toolkit. AI video models have dense training data on real video formats: GRWM lifestyle vlogs, sports stadium TikToks, news interview setups, nature documentary coverage, surveillance camera footage, dashcam video. When a prompt names one of these formats explicitly and follows its recognized conventions — the GRWM morning sequence opening with a curtain reveal, the stadium vlog cutting between front-camera selfie and rear-camera event coverage, the news interview with subject slightly off-lens — the model pulls its generation into alignment with its training on real footage in that format. The format name is a compression of hundreds of real-video behavioral conventions, and invoking it activates more of them than listing individual instructions would. Across all five techniques, the governing principle is the same: real video is defined by physical constraints and camera systems, not by content. Photorealism in AI video is a camera direction problem. The prompts in this gallery demonstrate what that looks like in practice — copy one, name the hardware, add the ban list, specify the physics, and name the format if it applies.

Seedance 2.0 AI video prompt: Reference character consistency: keep exact face, hairstyle, outfit, and body pr

Reference character consistency: keep exact face, hairstyle, outfit, and body proportions from the reference image. Realistic human appearance, cinematic lighting, grounded physics. Cinematic high-speed fight over an open ocean under a dark, stormy sky. Hyper-realistic water simulation, strong wind pressure bending waves, spray and mist trailing behind movement. 0–2s Wide aerial shot. The character flies at extreme speed above the ocean surface. Wind pressure creates long ripples and drifting spray. 2–4s A gigantic water elemental rises from a massive wave. Its body forms from swirling vortex currents, glowing green-blue eyes. It lunges forward, trying to grab her with a giant water arm. 4–6s She leans backward mid-air, narrowly dodging the grab, then counters with a powerful forward punch. The impact explodes into water spray and ripple shockwaves. 6–9s She lands and runs across the water surface at high speed. The creature swings another arm. She slides low to avoid it, then jumps and delivers a spinning kick. The water body deforms and splashes. 9–12s Slow-motion near-hit moment. A massive water hand almost grabs her. Droplets freeze mid-air. She twists sideways and escapes at the last second. 12–15s She accelerates forward, then turns and performs a final high-speed punch-and-kick combo. A massive water explosion erupts. The creature partially collapses into waves as mist fills the frame. Camera: dynamic tracking, whip pan, crash zoom, slow motion for near-miss moments. Lighting: storm clouds, subtle lightning, green-blue glow from the creature.

SUBJECTS: A human male sage @[image1] ENVIRONMENT: The entire space ultimately consists of a giant dragon’s body, but in the opening stage it appears as a weathered الأرض-like surface; the initial ground is a mix of compacted dust and eroded rock layers, with irregular surface detail and no scale characteristics; as the camera advances and scale shifts, the surface gradually reveals structured patterns and transitions into scale formations; the dragon’s body extends along the depth of the frame COLOR LOGIC: Overall cool gray-blue tones; avoid warm cinematic grading, avoid gold, avoid high saturation STYLE: cinematic realism, monumental scale, luxury campaign aesthetic, strong silhouette 0:00–0:03: The camera uses a 70mm medium telephoto, pushing in from a low position behind and to the side of the character with a slight lateral slide to create a compressed, oppressive composition; the character continues walking forward, his left foot landing on what appears to be ordinary weathered ground, producing weight-bearing contact and subtle dust particle disturbance; the surface consists of irregular compacted grains and wind-eroded textures with no scale features; particles shift and slide along the ground after compression; the body’s center of gravity moves forward into the next step; the cloak is pulled by a single-direction wind force, forming a stable flowing line; thin mist drifts slowly near the ground SFX: low wind, subtle particle friction 0:03–0:06: The camera quickly drops down close to the foot trajectory and slides upward along the body axis to the face; during the motion, ground particles detach, fall, decelerate, briefly suspend in the airflow, then slowly rise again; the camera settles on the face in a compressed composition; the character naturally lifts his head while walking SFX: particle rolling → airflow curl 0:06–0:09: The camera performs an approximately 120-degree semi-orbit around the character while slightly pulling back to open up space; the character continues walking; shoulder line and jaw angle subtly adjust with each step, forming a refined silhouette; the cloak forms a heavy, continuous flow line; the ground begins to show slight structured undulations, but still remains dominated by rock and sediment layers, only faintly suggesting organized structure rather than clear scales SFX: low-frequency air resonance 0:09–0:12: The camera rapidly pulls back and elevates, while the focal length gradually shifts to a wide angle to release spatial depth; the character becomes smaller in frame but continues walking; the ground gradually transforms into clear scale structures, transitioning from disorder to order; the overall curved surface reveals itself as the dragon’s back; the space becomes fully occupied by biological structure SFX: intensifying low-frequency vibration 0:12–0:15: The camera continues pulling back and rising into a wide angle, then rapidly decelerates while slightly pushing forward toward the dragon’s head to create visual pressure; the dragon’s body extends deeply into the frame and fully occupies the scene; the dragon’s head enters from the front side into a medium-close range, occupying the upper half of the frame; the head structure is elegant with clear flowing lines; scales are layered with natural transitions; the dragon’s eye is refined, with clear structure and multi-layered reflections without emitting light, conveying depth and life; the character remains at the lower center of the frame, extremely small but clearly visible, continuing to walk SFX: deep breathing sound

Seedance 2.0 AI video prompt: FORMAT: 15s / free rhythm / 1 MATCH CUT / CONTINUOUS MOVE UNTIL MATCH CUT + IMME

FORMAT: 15s / free rhythm / 1 MATCH CUT / CONTINUOUS MOVE UNTIL MATCH CUT + IMMEDIATE ACTION FROM FIRST FRAME SUBJECTS: A lone sword-bearing woman in weathered fur and leather fights a massive polar bear with desperate, two-handed survival movement. The same woman is later revealed at home in loose indoor clothes, where a VR headset appears only after the match cut and is pulled off in one clear motion. ENVIRONMENT: Frozen wilderness under hard daylight, wind dragging snow across blue-white ice, then a modest lived-in home reached through a precise visual match. Winter glare and visible breath give way to soft clutter, indoor daylight, and a faint game-lit glow. MOOD: Visceral survival tension snaps into grounded reality without breaking physical continuity. COLOR LOGIC: Naturalistic Film Print Emulation TIMELINE: 0:00-0:07: One unbroken handheld move, WS collapsing into MCU as the woman backpedals across the ice and the bear launches through blowing snow. The camera runs beside the leap at eye level, 28mm shifting to 35mm, slightly unstable and close enough to keep both bodies heavy and readable. The bear closes fast while she plants, recoils, and keeps the blade between them. SFX: (howling wind, boots grinding ice, low animal roar, cloth strain, blade cutting air, snow scrape). Hard winter sun side-lights the ice and throws sharp blue shadows. 0:07-0:11: Same unbroken move, no cut, tightening into a dead-on CU as the bear surges into the last inches, claws near her shoulders, jaws filling the frame edge. Right in the middle of the attack, a man's voice calls, Karla... then sharper, KARLA. She answers with a tired off, and on that reaction the world drops into slow motion. Snow drifts almost still, the bear hangs in its strike, and only she keeps moving at normal speed as the camera orbits into her face. Bored, not afraid, she drops the sword and brings both empty hands toward her temples in one smooth interrupt gesture. No headset, visor, or device is visible in the frozen world. Stay continuous until the match cut, keeping the same face size, hand height, head angle, lens distance, and clockwise drift. SFX: (cloth strain building to near impact, a man's voice calling Karla... KARLA, her tired off, then stretched wind fading toward silence). Hard winter sun catches the slowed snow around her face. 0:11-0:15: MATCH CUT. CU to MS. Seamless mid-motion transition as her rising hands cross the same screen position and the frozen close-up becomes the home interior with the same framing and clockwise drift. The motion continues uninterrupted, and now a VR headset is visibly strapped over her eyes for the first time. She grips both sides, pulls it fully off her face, and the camera opens into a medium shot as she drops it above her forehead and steps into a small living room in loose home clothes. The handheld orbit continues, revealing couch edges, scattered blankets, and cold window light as her posture falls into mild annoyance. She turns toward the voice, rolls her eyes upward, and says, What is it. 35mm natural lens, spherical. SFX: (headset strap stretch, plastic rub, quiet room tone, socked foot scrape, faint game audio, her breath settling, her dry voice saying What is it). Indoor daylight replaces the winter contrast.

Seedance 2.0 AI video prompt: Ultra-realistic Indian classroom street fight. Single continuous shot, no cuts.

Ultra-realistic Indian classroom street fight. Single continuous shot, no cuts. Raw handheld mobile footage with natural micro-jitters, slight rolling shutter, no stabilization. Documentary realism. Audio: No music. Only raw ambient sound-footsteps, desk friction, cloth movement, punches, fan hum, breathing, distant classroom noise. Lighting: Mixed warm + cool cinematic tones. Practical classroom lighting. Moving ceiling fan shadows. Visible dust particles in light shafts. MAIN CHARACTER (STRICT IDENTITY LOCK): Indian male (17-18), cold emotionless face matching reference image exactly. Black straight side-swept hair, brown almond eyes, sharp jawline, natural Indian skin tone, slim face. OUTFIT LOCK (NO CHANGE): Black school blazer, white shirt, black tie, black pants. No hoodie/tayers. Slightly fitted blazer, loose tie. Must remain identical throughout OPPONENT RULES: All opponents have unique faces. No resemblance to protagonist. ACTION: 0-2s: Protagonist grabs Opponent A from behind one-hand head slam onto desk. Books/pencils burst outward. Desk drags with friction. A collapses limp. 2-5s: Grabs Opponent B punch to abdomen B folds immediate head slam into desk. Head pressed briefly to tiles. Camera stabilizes briefly. B falls face-down. Heavy breathing, visible sweat. 5-6s: Blazer shifts slightly, shirt + tie visible (tie displaced). C & D enter from back, split and flank. Only footsteps, fan hum. 6-8s: Chaotic fight. D attacks with rapid punches. Protagonist stumbles slightly. C grabs blazer, pulls back. Knee to abdomen. Body bends forward. Heavy breathing. Fan shadows moving. 8-10s: Headbutt to D D staggers. Protagonist grabs C collar efficiently. 10-12s (BOARD IMPACT): Protagonist drives C backward into green chalkboard. Back hits first, head snaps slightly. Chalk dust bursts outward, fully visible in warm light. Chalk tray rattles, pieces fall. Micro slow-motion (0.2-0.3s) at impact back to real-time. C loses tension, slides down leaving chalk smears, collapses. Dust drifts. 12-13s: D charges protagonist sidesteps rotational punch to cheekbone. Sweat arcs mid-air. D crashes into desks. Massive collapse, books scatter. D motionless, hand drops. Protagonist barely standing, heavy breathing. 13-14s: Medium-close. Chest rising. Slow turn to camera. Cold eye contact through glasses. Wipes blood from nose (visible smear). Removes glasses throws aside, spinning impact. 14-15s (FINAL): Ground-level wide shot. Slight edge blur, warm tone, golden dust. Protagonist without glasses removes blazer → throws onto opponent. White shirt + black tie only. Tie loose, slight movement. Walks to exit. Blazer settles. Silent classroom. Dust floating. Fade to black.

Seedance 2.0 AI video prompt: Image1 character as the subject a male, black oversized hoodie, black oversized

Image1 character as the subject a male, black oversized hoodie, black oversized baggy jeans, black sneakers. Radiation-enhanced superhuman speed and strength, awakened orange-yellow glowing pupils, veins bulging. intense aura. Scene: Abandoned Chicago high-rise ruins, exposed rebar, cracked concrete, suspended derelict vehicles, rusted steel walkways, mutant vines, broken glass shards, dim moonlight, glitching neon flicker, thick fog, extreme high-altitude tension. One-take continuous shot, 15 seconds, cinematic, 120fps high frame rate, ultra-realistic dark American post-apocalyptic sci-fi thriller, desaturated cold gray palette with orange-yellow accent. 0-3s: Low angle wide shot character stands at edge of cracked floor slab, panting after clearing small enemies. orange-vellow pupils coldly scanning ruins. Building violently shakes. Upgraded Giant Tracker bursts through fog, slams floor with massive claws, concrete explodes. Handheld slight tremor, wide angle oppressive framing. 3-6s: Giant Tracker sweeps massive claw character side-dashes at extreme speed with strong motion blur afterimage. Claw shatters steel column, dust blast. Monster strikes repeatedly, character serpentine-sprints through debris gaps at high speed. Ground-level chase track shot, full motion blur. 6-8s: Character kicks off suspended vehicle in mid-air, rotates at high speed, launches spinning kick at monster's head. Monster roars, snapping jaw barely grazes his sneaker. Hitchcock spiral orbit shot, high-altitude danger. 8-10s: Character lands and explodes forward -three rapid heavy punches to monster's chest and jaw, each blow forcing it back, bone-crushing thuds. Low sweep kick destabilizes monster's footing. Medium close-up tracking, 120fps slow-motion captures fist impact force. 10-13s: Character grabs monster arm, reverse-twists exoskeleton cracks and bursts. Leaps onto monster's back, legs lock around throat, both arms pry apart neck armor. Monster thrashes violently, character holds firm. Back overhead shot with fast lateral cuts, strong motion blur afterimage. 13-15s: Character vaults off, charges full-force punch into monster's split skull jaw kinetic force detonates through the head, monster collapses, shockwave shatters surrounding vines and glass. Final freeze frame: character stands atop monster corpse, orange-yellow

Seedance 2.0 AI video prompt: 15-second, no-dialogue, fully immersive high-speed brutal fight short film.

Two

15-second, no-dialogue, fully immersive high-speed brutal fight short film. Two characters @ image1 and @ image2 . Live-action realistic style. Close-quarters, high-speed chaotic melee combat with no rules, fully out-of-control brawling. No flashy choreography—pure raw. Continuous exchange of punches, kicks, blocks, body checks, throws, and grappling entanglements. Ultra-realistic impact. Exaggerated yet physically grounded motion speed. Entire sequence in 120 FPS with no slow motion. Rapid dashes, evasions, and high-speed limb swings. Extreme motion blur and speed trails amplify intensity while preserving realistic weight and physics. Master-level cinematic camera movement: A professional film camera stays tightly locked to the action, flying dynamically with the fighters—high-speed dives, sudden stops, whip pans, and full 360° orbital tracking. The camera moves in sync with punches, kicks, and evasions, with responsive shake and displacement. One continuous take, no cuts. Sharp push-pull tracking, sudden directional shifts, fully integrated into the fight for a face-to-face immersive combat perspective. Smooth, zero-latency motion. Master-level precision editing: 0–3 seconds: ultra-fast cuts to enter combat. 3–10 seconds: seamless action continuity with zero stutter or breaks. 10–15 seconds: dense, violent rapid-cut climax. Editing rhythm escalates with combat intensity. Fast intercutting enhances chaos. Precise beat timing, no redundant shots. Constant suffocating high-speed pacing. Environment Setting Modern open-plan office with desks, computers, cabinets, printers, and swivel chairs; realistic layout with cool overhead lighting and side backlight for strong contrast. Slightly cramped space for added tension; no glass-breaking. Environmental Interaction & Constraints: Combat affects the space only through natural, physically justified collisions—papers scatter from air movement (not attacks), chairs slide or topple on impact, desks shift slightly without deliberate damage, and screens remain intact with minor vibration. No unnecessary or intentional destruction of equipment. Body-driven combat: grappling, pinning, counter-throws, and realistic slams with subtle impact cracks/vibrations. Natural airflow—light paper/dust movement only, no exaggeration. Final Action Design Final 2–3s: both throw simultaneous close-range punches, freezing inches before impact—fully tensed, heavy breathing, locked in perfect deadlock.

Seedance 2.0 AI video prompt: Cinematic continuous side-scrolling sequence, 16:9, 15 seconds. One unbroken sho

Cinematic continuous side-scrolling sequence, 16:9, 15 seconds. One unbroken shot — the camera tracks smoothly from left to right without cutting, following a single person through their entire life as the environment transforms seamlessly around them. Opens in a hospital delivery room — a newborn baby crying, bright sterile lights, doctors and a mother. The baby crawls right and the world transforms. A colorful chaotic kindergarten — finger paintings on walls, blocks scattered everywhere, the child running and laughing loudly, falling over, getting back up. Keeps moving right — a school playground, kid sprinting full speed chasing friends, shouting, scraped knees, backpack bouncing. Right again — a teenage bedroom, loud music thumping, jumping on the bed, laughing with friends, energy everywhere. Right — a college campus, rushing between classes, coffee in hand, stressed, alive. Right — a first job, city streets, rushing through crowds, briefcase, purpose and chaos. Right — a wedding, spinning a partner, confetti exploding, crowd cheering. Right — a park with young children running around them, chaotic and beautiful. Right — a quieter home, slower pace, reading glasses, warm light. Right — very old now, walking slowly with a cane, white hair, soft smile, grandchildren pulling at their hands. The environment dims gently, pace slows, the figure sits in a chair by a window, golden light, closes their eyes. Sound design matches every stage — hospital beeps, children screaming and laughing, playground chaos, thumping music, crowd cheering, city noise, birds, silence. Photorealistic, IMAX cinematic quality, ultra sharp, vivid colors throughout each life stage, seamless environmental transitions, deeply emotional and cinematic

Seedance 2.0 AI video prompt: Character tone: high-end romantic comedy, deadpan flirtation, over-serious male

Character tone: high-end romantic comedy, deadpan flirtation, over-serious male lead, quick-witted female lead, cinematic realism, sweet-chaotic chemistry, every frame like a poster Male lead: bespoke black suit, white shirt collar slightly open, handsome and severe, powerful aura, trying very hard to look cold and dominant, but secretly nervous and flustered, tiny tells betray him: slightly crooked tie, tight jaw, faintly trembling fingertips Female lead [@ Image1]: fitted slip dress / refined Chanel-inspired set, long hair slightly messy, elegant and soft-looking but emotionally sharper than him, stubborn, dryly funny, outwardly cornered for a moment, then visibly unimpressed, holding back laughter Action + expression changes: the male lead forcefully steps in for a dramatic wall-pin pose, one hand braced on the wall, closing the distance too seriously, trying to look intense; his expression starts cold but gradually cracks into restrained embarrassment the female lead [@ Image1] steps back once, eyes widening, then notices his crooked tie and trembling hand; her expression changes from guarded resistance to deadpan disbelief and almost-laughing annoyance their noses nearly touch, breathing overlaps, the tension becomes playful and absurd instead of painful the male lead lightly lifts her chin, trying to recover his cool image; the female lead stares at him like she is watching someone forget his own script Dialogue: Male lead: Are you done yet? Female lead [@ Image1]: Fix your tie first. Male lead: I am being serious. Female lead [@ Image1]: Then stop stepping on my heel. Male lead: ...That was deliberate. Female lead [@ Image1]: Your shaking hand says otherwise.

Seedance 2.0 AI video prompt: Create a photorealistic ultra-cinematic 15-second first-person POV action scene.

Create a photorealistic ultra-cinematic 15-second first-person POV action scene. We are riding on the back of a massive black dragon racing through a storm-torn canyon at night. Both hands remain visible in frame, gripping the dragon’s wet armored neck ridges like living reins. Wind lashes hard, rain cuts sideways, lightning reveals jagged cliffs and abyss drops below. The tone is epic, intense, and premium blockbuster fantasy realism. 0-3s: first-person POV low over the dragon’s neck, both hands gripping tightly as the dragon surges through the canyon at high speed, black scales slick with rain, breath vapor and embers trailing from its mouth, cliffs rushing by on both sides. 3-6s: ahead in the storm, a squad of armored aerial hunters appears and attacks, launching multiple flaming bolt-spears directly toward us. The projectiles race through the rain toward camera while the dragon banks violently between cliff walls. 6-9s: at the last possible instant, one hand releases the dragon and rises into frame with a sharp command gesture. Time stops instantly. All incoming bolt-spears freeze in mid-air just in front of the POV, along with rain droplets, sparks, flames, and drifting mist, suspended in a rippling invisible field. 9-12s: the raised hand slowly twists sideways. The frozen incoming projectiles rotate in the air one by one until all of them are now pointing back toward the attacking riders. The enemies remain moving in the distance, but the captured weapons are fully controlled and aimed at them. 12-15s: with a second precise hand gesture, time snaps forward only for the redirected projectiles, sending the entire suspended volley back at extreme speed into the attacking aerial hunters. The returning flaming bolt-spears slam into the enemy formation in the storm as the dragon drives forward through the chaos. End on a powerful forward-moving POV frame with one hand returning to grip the dragon.

Seedance 2.0 AI video prompt: Theme: Classic fight between young and beautiful men and women - modern short dr

Theme: Classic fight between young and beautiful men and women - modern short drama Length: 15 seconds Aspect ratio: 16:9 15-second modern short drama, luxurious CEO office scene, cool color tone and high-end feel. 0-3 seconds: Medium shot. A young man in a bespoke suit sits in a leather executive chair, his face cold and his eyes indifferent. A beautiful girl in a white dress stands before him, her eyes red and brimming with tears. The man speaks, "I gave you a chance," his voice low and cold. 4-8 seconds: The girl's voice trembles, "What chance did you give me? You never trusted me from the beginning!" She turns to leave, but the man suddenly stands up and grabs her wrist. The camera follows their struggle, forcing the girl to turn back. They are extremely close, their breaths mingling. A flicker of hesitation flashes in the girl's eyes, quickly replaced by stubbornness. 9-12 seconds: The man stares into the girl's eyes, his voice low, "So, are you still going to leave?" The girl bites her lip, "Go...where else can I go?" She cruelly shakes off his hand and turns to run. The man reaches out to grab her but grasps only air, his hand frozen in mid-air. 13-15 seconds: The camera follows the girl as she rushes out of the office. She wiped away tears in front of the elevator, but the tears wouldn't stop flowing. As the elevator doors slowly closed, she crouched down, buried her face in her arms, and sobbed. The final frame of the scene was a close-up of the elevator numbers dropping from 15, accompanied by a low, sorrowful piano background music. A 15-second modern campus short drama, set in a rainy night classroom corridor, with a cool blue-toned rainy night atmosphere. 0-3 seconds: Fixed shot. Heavy rain pours outside the classroom, raindrops patter against the windowpanes. A handsome boy in a white shirt stands in the shadows of the corridor, pacing nervously back and forth, clutching a pink letter. In the distance, a pretty girl with a high ponytail walks slowly towards them. The camera shakes slightly with each step she takes to create tension. 4-8 seconds: The girl approaches the boy. He takes a few deep breaths, raises his hand as if to hand her the letter, then pulls it back. The girl stops in front of him, tilts her head, and smiles, "Did you need something?" The boy stammers, "I...I..." Before he can finish, the bell rings, startling them both. The corridor lights flicker. 9-12 seconds: The girl glances at the time, "Class is about to start, let's talk about it next time," and turns to leave. The boy musters his courage, "Wait!" He grabs her wrist, and they both freeze. Only the sound of rain and their breathing fills the corridor. The boy looks down, his cheeks flushed, "This letter." "...Here you go," the girl paused, looking down at the letter; 13-15 seconds: the camera slowly zooms in on the girl's face, a sweet smile slowly spreading across her lips, starlight twinkling in her eyes. The final shot freezes on the moment she takes the letter—a ray of sunlight suddenly streams in through the rain-soaked window, illuminating the two of them, the scene ending amidst a gentle piano melody and the soft sound of rain. 15-second modern short drama, upscale restaurant scene, warm gold tones contrasting with a cold mood. 0-3 seconds: Medium shot, fixed angle. A beautiful girl is celebrating her birthday with friends, raising a glass and laughing joyfully. Suddenly, the restaurant door is pushed open—her most trusted best friend walks in arm-in-arm with her boyfriend. The laughter stops abruptly, and the girl's hand, holding the glass, freezes in mid-air, her smile captivated. 4-8 seconds: The best friend's expression becomes awkward upon seeing the girl. She tries to pull her boyfriend to move to another seat, but he shakes her hand and walks straight to the girl, placing a gift box in front of her and saying "Happy Birthday." The best friend's face turns pale. The girl looks at her boyfriend, then at her best friend, finally focusing on their hands—the best friend is wearing the same bracelet her boyfriend said he "lost" a few days ago. 9-12 seconds: The girl doesn't speak, but quietly picks up her glass and drinks it all in one gulp. Then she stands up and walks to her best friend. Everyone holds their breath, waiting. The girl smiles and slowly pours the wine onto her best friend's limited-edition bag. The best friend's eyes widen, as if she's about to scream. The girl holds her shoulder; 13-15 seconds: The girl leans close to her best friend's ear and whispers, "Next time you're cheating, remember to hide what you shouldn't show first." The best friend turns pale, and the girl turns and leaves gracefully. The camera pulls back to show the best friend standing there stiffly, her boyfriend full of regret, ending with a low background music and whispers around them.

FORMAT: 15s / single continuous impossible camera move / no dialogue STYLE: Dense Southeast Asian night market, wet stone, steam and fire, orange lantern light, photorealistic ground-to-aerial cinematic 8K Shot 01 (0:00–2:00): Camera starts at ankle level. Forest of feet — sandals, bare feet, flip flops on glistening wet stone. Market sounds overwhelming. Camera weaves between legs like water. Audio: Wet footsteps, distant vendor calls, sizzling. Shot 02 (2:00–3:30): Camera rises slowly past steaming woks. A gas flame bursts at exact eye level — camera briefly engulfed in orange fire then emerges through it, uncut. Audio: Gas flame WHOMP. Wok sizzle. Shot 03 (3:30–5:00): Camera weaves through hanging lanterns at mid-height, skimming past them like a moth navigating. Red and orange light strobing across the lens. Audio: Lantern chains clinking. Crowd murmur. Shot 04 (5:00–6:30): Camera dips suddenly under a low table. A child has fallen asleep on a bag under here. Quiet pocket of stillness inside the chaos. Camera lingers one beat. Audio: Noise muffled. Child's quiet breathing. Shot 05 (6:30–8:00): Camera rises back up through the smoke of a charcoal grill — lens briefly obscured by smoke, then emerges above the stall canopy level. Audio: Charcoal crackle. Smoke hiss. Shot 06 (8:00–10:00): Camera continues rising — now above rooftop level. The market reveals itself as an endless orange lantern sea stretching to the horizon. Camera tilts slowly, taking in the full scale. Audio: Market sounds fading to a low ambient hum. Shot 07 (10:00–12:00): Camera begins descending back down — targeting a single stall at the market edge. A lone vendor counting coins. Everything closing around them. Audio: Distant vendor calls dying out one by one. Shot 08 (12:00–13:30): Camera settles at the vendor's hands. Close on coins being stacked methodically. One lantern swaying above them. Audio: Coin clinks. Wind in the lantern. Shot 09 (13:30–15:00): Camera tilts up slowly to the vendor's face. They look up — directly into the camera. Hold. Fade to black. Audio: Last coin placed. Silence. STYLE NOTES: Ground level feels claustrophobic and overwhelming. The aerial shot is the emotional exhale. Warm orange and red throughout — deepening to amber at the end. Wet stone reflections throughout. Real fire and real smoke, no CG substitutes. 8K.

Seedance 2.0 AI video prompt: A young woman matching the exact facial features, hairstyle, and identity of the

A young woman matching the exact facial features, hairstyle, and identity of the reference image (consistent character throughout all shots) is filming a “Get Ready With Me” style video in a softly lit, modern bedroom with a vanity mirror and warm ambient lighting. Opening scene (casual look): She is wearing comfortable knee-length fitted shorts (extending just over the knees) and a simple sleeveless crop top. Natural look, no makeup. She smiles at the camera and casually greets, creating a relaxed, vlog-style tone. Sequence of shots (fast-paced, smooth cuts, ~10–15 shots total): Close-up of her tying her hair into a loose bun Applying face wash at the sink, soft water splashes, natural lighting Gently patting her face dry with a towel Skincare routine (serum, moisturizer) in front of mirror Sitting at vanity applying makeup step-by-step (foundation, eyeliner, lipstick) She occasionally looks into the camera and subtly explains each step (natural lip movement, expressive gestures) Insert aesthetic close-ups: makeup brushes, mirror reflections, soft bokeh lighting Tone & style: Cinematic but vlog-inspired, smooth camera motion, shallow depth of field, soft highlights, natural skin texture, realistic lighting. Maintain facial consistency across all shots. Transition: Quick montage with rhythmic cuts as she completes her look. Final reveal scene: She steps back into frame fully dressed in an attractive yet modest red outfit (elegant, tasteful, non-revealing, flattering silhouette). Hair styled neatly, confident posture. She does a subtle twirl or pose, smiles warmly at the camera, and gives a final confident look. Ending shot: Soft close-up with warm lighting, slight camera push-in, she looks directly into the lens with a satisfied expression. Technical tags: 1080p, cinematic lighting, consistent character identity, smooth transitions, natural motion, realistic skin tones, high detail, depth of field, vlog + cinematic hybrid style

Seedance 2.0 AI video prompt: @Image1 is the main character — maintain consistent facial features and body typ

@Image1 is the main character — maintain consistent facial features and body type throughout. Cinematic time-freeze short film, 15 seconds, ultra-realistic, Arri Alexa Mini shooting texture, 50mm lens, natural daylight hard shadows, shallow depth of field. [0:00–0:03] Busy city sidewalk, normal time flow. Steadicam front-facing medium shot tracking: the main character wearing a dark jacket walks confidently through the crowd. Pedestrians walk, check phones, chat; a flock of pigeons flies across the bright sunny sky in the distance. As he walks, he raises his right hand — and snaps his fingers. [0:03–0:06] The instant of the snap — a powerful white spherical shockwave bursts from his fingertips, carrying visible air distortion and light refraction, spreading rapidly in all directions. It sweeps across the ground, kicking up dust and fallen leaves that freeze mid-air instantly. It sweeps over pedestrians — each one freezes mid-stride. It sweeps across the sky — every pigeon locks in a different wing-spread pose, hovering motionless in mid-air. The shockwave ripples a streak of light across distant building windows. After the shockwave vanishes — absolute silence. Subtle time-freeze particles float in the air, gently shimmering in the sunlight. [0:06–0:09] In the silence, only the main character's footsteps echo crisply. Camera flies backward ahead of him — he walks toward the lens, casually strolling through the frozen crowd, looking around this still world with a smug expression. He looks up and sees a pigeon hovering above, reaches out and gently strokes its feathers. The pigeon doesn't move at all, like a sculpture, but lets out a deep coo that reverberates through the silence. Satisfied, he keeps walking. [0:09–0:11] The main character walks up to a frozen young woman — she's wearing a flowing red dress, her long hair caught by the wind and frozen mid-air in an elegant arc, one hand locked in a hair-tucking pose, expression graceful. He stops, hands in pockets, tilts his head and admires her. He comes around to face her, gives a slight nod with a corner smile, and whispers in Turkish: "harika." [0:11–0:15] The main character turns around, pulls a cigarette from his jacket's inner pocket and places it between his lips, flicks open a lighter — click — lights it, takes a drag and slowly exhales smoke — the smoke is the only thing moving in the entire frozen world. With the cigarette in his mouth he turns to face the camera, smirks, raises his right hand and snaps again — a second shockwave, even more powerful, bursts outward from his fingertips in reverse, carrying air distortion and light refraction, spreading at high speed. Everything it touches instantly unfreezes: pedestrians resume walking, pigeons scatter flapping, the woman in the red dress continues tucking her hair and walks on completely unaware, fallen leaves finally land. City noise naturally returns. The main character, cigarette dangling, calmly turns and walks on. Camera slowly rises and pulls back into an overhead wide shot — his silhouette blends into the city. Fade to black. Sound design: City noise → finger snap triggers a low-frequency explosive shockwave rumble radiating outward from near to far → absolute silence → footsteps echoing → pigeon cooing → whispered Turkish "harika" → lighter click → inhale and exhale of smoke → second finger snap → stronger reverse shockwave explosion rumble → city noise naturally restored.

[Style] Hollywood Haute Couture Fantasy blockbuster, 8K ultra-clear, Photorealistic, High-fashion Editorial Style, Unreal Engine 5 fluid rendering, visual illusion. [Duration] 15 seconds. [Scene] An endless, real-life Salar de Uyuni (Sky Mirror) salt flat. The sky is filled with oppressive dark clouds, and the ground perfectly reflects everything like a mirror, with the overall picture presenting a minimalist, cool tone. [00:00-00:05] Shot 1: Haute Couture Entrance and Porcelain Skin. Camera position: Extremely low-angle upward shot, ultra-telephoto lens zoom-in. Action: An Asian female model with a highly recognizable, high-fashion face walks coolly on the water surface. Effect: She is wearing not fabric, but a long dress made of flowing, real Liquid Blue-and-White Porcelain. As she walks, the skirt makes a crisp collision sound like real ceramic, with a flowing luster on the surface. The traditional blue-and-white patterns move across the white porcelain-textured skirt as if alive. [00:05-00:10] Shot 2: Physical Shattering and Ink-wash Descent. Camera position: Extreme close-up of the face, focus rapidly pulls back. Action: The model suddenly stops, stares coldly at the camera, and snaps her fingers crisply. Effect: The moment the fingers snap, her blue-and-white porcelain dress does not fall, but instantly explodes into thousands of extremely photorealistic Ink-wash Swallows. These swallows carry real water droplets and ink marks, dragging black fluid afterimages in the air, spinning frantically around her. [00:10-00:15] Shot 3: Dimensional Dissolution and Abyss Reflection. Camera position: High-altitude overhead shot, camera rapidly rotates and descends. Action: The swarm of ink-wash swallows plunges into the mirrored lake water beneath the model's feet. Effect: The surface tension of the originally solid salt lake instantly disappears. The entire extremely realistic world begins to violently bleed and dissolve like concentrated ink dropped into clear water. The real dark clouds and the model's figure transform entirely into an extremely grand 3D Fluid Ink Vortex, completely swallowing the camera into a black and white interwoven abyss.

Seedance 2.0 AI video prompt: Use the uploaded 2x2 @image1 grid as a contact sheet of memories, not as a liter

Use the uploaded 2x2 @image1 grid as a contact sheet of memories, not as a literal grid. Transform the individual photos into one cohesive nostalgic music video sequence about the same young man and young woman falling in love through candid iPhone moments. Create a 12–15 second emotional indie-pop MV montage. Preserve the same couple’s faces, age, hair, clothing style, and relationship energy across all shots. The video should feel like real iPhone footage from a private camera roll: imperfect, intimate, handheld, slightly shaky, amateur framing, accidental zooms, motion blur, blown highlights, soft grain, muted vintage colors, warm sunset tones, and tiny focus mistakes. The sequence should flow like fragmented memories: Editing style: cohesive music video montage, romantic but not cheesy, dreamy memory fragments, rhythmic cuts synced to an imagined melancholic indie-pop song, natural handheld movement, soft whip-pan transitions, phone-camera flash transitions, match cuts between similar poses from different photos. Visual style: candid iPhone photography, 2010s camera roll nostalgia, vintage emotional aesthetic, amateur home-video realism, soft film grain, subtle VHS-like imperfection, warm faded colors, realistic skin texture, natural expressions. Important rules: do not show the original grid layout, do not create a collage, do not add text, do not add watermarks, do not introduce new people, do not make them look like fashion models, do not over-cinematize, do not make the camera too smooth, do not change their identities between shots. Keep it intimate, messy, romantic, and emotionally real.

Seedance 2.0 AI video prompt: A hyper-realistic sequence showing a photorealistic Formula 1 car emerging from

A hyper-realistic sequence showing a photorealistic Formula 1 car emerging from a detailed pencil sketch on a cluttered architect's desk.", "camera": { "type": "Slow Arc Shot", "lens": "100mm Macro Prime", "aperture": "f/2.8", "position": "Starts as an extreme close-up on the pencil drawing, then slowly pulls back and arcs around the car as it transitions to a 3D object.", "environment": { "location": "A designer's or engineer's drafting table.", "details": "The surface is a sheet of high-quality, slightly textured white paper covered in technical blueprints and schematics of the car. Several graphite and colored pencils, an eraser, and a compass are scattered around the main drawing. The background is a dark, out-of-focus workshop." }, "lighting": { "primary_source": { "type": "Architect's Desk Lamp", "position": "Just out of frame, to the top left.", "fill_light": { "type": "Soft Ambient Light", "source": "From the out-of-focus workshop background.", "color": "Cool, bluish tint.", "intensity": "Low, to gently lift the darkest shadows and add depth to the scene." } }, "main_subject": { "name": "The Formula 1 Car", "initial_state": "A masterful, hyper-realistic pencil drawing. The lines are precise, the shading is smooth, and there's a sense of depth even in 2D.", "final_state": "A fully realized, photorealistic 3D model of the Red Bull F1 car. The materials are perfect: glossy carbon fiber, metallic paint with livery, and soft-sheen rubber tires.", "animation": { "0.0s - 2.0s": "The camera is focused on the drawing. A photorealistic hand enters the frame with an eraser and gently removes a few faint construction lines from the sketch.", "2.0s - 7.0s": "A magical 'wave' of transformation begins at the front wing of the car and sweeps backward. As it passes, the 2D pencil lines and shading seamlessly transform into 3D, textured materials. The drawn debris on the track lifts up and becomes real, floating for a moment before settling.", "7.0s - 8.0-10s": "The transformation is complete. The car is now fully 3D, sitting on the paper as if it's a real, miniature vehicle. The engine subtly whirs to life, and the exhaust vents emit a tiny, realistic heat haze." } }, "audio_design": { "style": "Transformative, from quiet and organic to powerful and cinematic.", "timeline": [ { "time": "0.0s - 2.0s", "sound": "The soft, realistic sound of a pencil's graphite scratching on paper, and the gentle 'shhh' of an eraser." }, { "time": "2.0s - 7.0s", "sound": "A low, magical humming and shimmering sound that grows in intensity and pitch as the transformation wave moves across the car. This is layered with an epic, swelling orchestral score." }, { "time": "7.0s - 10s", "sound": "

Seedance 2.0 AI video prompt: Character A highly trained female operative in a sleek black tactical suit, calm

Character A highly trained female operative in a sleek black tactical suit, calm, focused, controlled breathing, sharp eyes scanning constantly. Dual pistols in hand, precise and efficient. 0-3s Camera starts inside a luxury bank. Slow push through a tense standoff—security guards aiming, civilians frozen. She stands in the center, perfectly still. 3-5s Without warning—everything explodes into motion. She steps forward. Rapid precise shots—each movement minimal, controlled, no wasted motion. 5-7s Camera locks close to her, orbiting as she moves through guards. She uses environment—pillars, desks—flowing movement, never stopping. 7-9s Transition through glass reflections. She slides across polished floor, firing mid-motion, bullets impacting surfaces around her. 9-11s Tight corridor sequence. Close-quarters engagement, ultra fast reactions, micro head turns, instant targeting. 11-13s Massive push forward into main hall. Camera pulls wide—chaos everywhere. She moves through it like it’s slowed down. 13-15s Final moment. She exits through shattered glass doors into daylight. Doesn’t look back. Behind her—delayed impacts, everything collapses. Camera freezes on her walking away. Continuous tracking shot, tight proximity + dynamic orbit + whip transitions masked by movement Luxury bank interior, marble, glass, reflections, daylight contrast Ultra realistic, hollywood heist action, john wick gun-fu, precise choreography, high tension, fluid motion, cinematic lighting, no distortion, no stretching

Seedance 2.0 AI video prompt: SUBJECTS:
A dominant male leader, short hair, light stubble, strong jawline, pie

SUBJECTS: A dominant male leader, short hair, light stubble, strong jawline, piercing gaze, broad shoulders, muscular build. Wearing a white structured ceremonial war robe with layered panels and split drapes, wide reinforced belt, central rigid chest plate, and a heavy asymmetrical black mechanical shoulder armor. Carries a divine lightning spear, long ceremonial polearm, polished silver and ivory metal, energy conductor structure, subtle lightning gathering along the tip, controlled electric arcs. ENVIRONMENT: A colossal sacred western mythic megastructure city. Architecture inspired by divine western civilization: cathedral-scale sanctuaries, monumental marble structures, radiant statues of angelic guardians and noble deities. Massive layered city with canals, bridges, terraces. COLOR & LIGHTING: soft diffused lighting, gentle warm highlights, soft natural glow, subtle teal water tones, muted greens, warm natural skin tones, soft shadows. STYLE: live-action cinematography, continuous physical camera motion, natural inertia, no CGI feel, real textures, real atmospheric perspective, subtle imperfections, no stylization artifacts. TIMELINE: 0:00–0:03 Camera dives from high altitude with strong forward momentum, slightly tilting and accelerating. A massive pale moon sits far in the background, softly diffused through clouds. Two colossal sacred statues emerge through mist, designed as noble angelic guardians with refined, harmonious facial features. Camera threads precisely between them, maintaining forward inertia. 0:03–0:06 Camera continues seamlessly into a downward spiral, following architectural curvature without any cut. A majestic dragon is perched across a grand stone bridge ahead. As the camera approaches, the dragon turns and releases a wide, flowing stream of fire. The flame expands across the path like a luminous veil. Camera rolls slightly and glides beneath the fire in one continuous motion, entering the shadow of the bridge, then following the underside structure forward, naturally transitioning toward the canal below. 0:06–0:09 Maintaining the same motion path, the camera exits the bridge shadow and drops closer to the canal surface. Water reflects soft firelight and surrounding architecture. Camera continues forward and passes under a lower arch. Reveals a breathtaking scene: marble ruins, sculpted pillars, statues softened, light rays diffusing through water. Camera follows forward trajectory underwater, then rises naturally toward a fountain outlet and bursts upward, maintaining direction and momentum. 0:09–0:12 Camera flips with inertia and sweeps across a vast ceremonial plaza. Hundreds of soldiers stand in formation. The leader stands facing the army. He steps forward forcefully, lifts the lightning spear high, and channels energy upward. Soldiers respond in unison with a powerful synchronized roar, raising weapons in a wave-like motion. Subtle electrical particles begin to gather in the air, dust and embers lifting upward as if drawn by energy. Camera weaves through the formation at low height. 0:12–0:15 Camera ascends along a spiral path toward the summit platform, continuous motion. Fire braziers ignite sequentially along the ascent. As the camera rises, a massive dragon suddenly cuts through the fog layer from the side, flying close across the frame, wings displacing mist and light. The camera reaches the summit. The leader stands elevated, facing the army, spear raised high. The sky above compresses with dense clouds, energy concentrates, and a massive lightning strike descends from the sky, connecting directly to the spear in a controlled, continuous flow. The entire scene is illuminated by the lightning, soldiers below erupt again in unified response. In the distance, another dragon circles slowly. Camera pulls back smoothly, revealing full scale: layered fog, fire accents, and the moon remaining as a subtle background.

Seedance 2.0 AI video prompt: Ultra-realistic 6-second miniature diorama video: A tiny realistic human family

Ultra-realistic 6-second miniature diorama video: A tiny realistic human family stands worried on a miniature suburban street as their small kitten is stuck on a high rooftop. A giant realistic human hand with natural skin texture and gentle fingers slowly reaches down from above, carefully scoops the kitten into its palm, and safely places it back with the joyful family who wave in gratitude. Cinematic tilt-shift lighting, warm golden hour glow, highly detailed tiny people and textures, smooth slow-motion rescue, heartwarming and wholesome. 6-second hyper-realistic AI video in miniature scale: Miniature realistic humans desperately try to save their flooded vegetable garden as rainwater turns it into a tiny lake. A giant realistic human hand enters from the sky holding a large leaf like a dam, gently redirects the water flow away from the plants, then pats the soil dry. Detailed wet textures, natural hand veins and subtle finger adjustments, cinematic close-up, satisfying resolution with relieved tiny gardeners smiling. Cinematic 6-second miniature scene: Tiny realistic office workers are trapped under a collapsed miniature bookshelf during a 'quake'. A giant realistic human hand with lifelike skin and careful grip slowly lifts the heavy books one by one, creating a safe path, then helps the last tiny person stand up. Warm indoor lighting, extreme detail on miniature books and clothing, gentle compassionate movement, uplifting and creative help. 6-second realistic miniature diorama: A tiny realistic adventurer's hot air balloon is tangled in miniature tree branches high above a park. A giant realistic human hand reaches in delicately, untangles the ropes with precise finger movements, then gives the balloon a soft upward nudge to set it free as the tiny pilot cheers. Soft natural daylight, detailed fabric and rope textures, whimsical yet realistic, smooth satisfying release. Hyper-detailed 6-second AI video: A miniature realistic child sits sadly beside their broken tiny bicycle on a cobblestone street, chain off and wheel bent. A giant realistic human hand appears, carefully straightens the wheel, reattaches the chain with gentle precision, then gives the bike a light push to test it as the child hops on happily. Street-level cinematic angle, realistic hand skin and subtle shadows, creative everyday kindness, warm and positive ending.

FORMAT: 15s / 145 BPM / 15 SHOTS / beat-synced routine SUBJECT: @[image1]. WARDROBE: Sleep tee and lounge shorts at home. Tailored jacket, fitted top, trousers, and lace-up shoes outside. ENVIRONMENT: Tiny apartment, bright fridge glow, rain-dusted hallway, chrome metro, clean office, then a bedroom in cool window light. Everything feels glossy and lived-in. MOOD: Late-for-work panic, clipped momentum, breathless urgency, then an exhausted exhale. MUSIC: Fast percussive electro-pop COLOR LOGIC: Hyperreal Pop Look STYLE: Ultra-Realistic. LOGIC RULE: Keep logical consistency in wardrobe, props, locations, and action continuity across all shots. SHOT 1: ECU, 85mm push-in / 06:50 on the phone screen as it shakes on rumpled sheets. / SFX: alarm, sheet rustle. SHOT 2: WS, 35mm handheld jolt / Rhythmic cut into her jolting upright through side light, throwing the blanket aside, and planting her feet on the floor in one rushed motion, still in a soft sleep tee and lounge shorts. / SFX: mattress bounce, blanket whip, sharp breath. SHOT 3: MCU, 50mm slide / Cut on action into face wash at the sink, droplets catching the top light. / SFX: faucet rush, water slap. SHOT 4: Insert shot, 85mm rack focus / Match cut into the toothbrush held at a natural forward brushing angle against the front teeth, hand relaxed and upright, mint foam and mirror eye. / SFX: bristle scrape, sink drip. SHOT 5: Interior fridge view, 24mm wide / Object pass into the camera inside the fridge looking out as the door snaps open and her hand darts in, blue fridge light framing a hurried grab for breakfast ingredients. / SFX: fridge hum, bottle clink, shelf rattle. SHOT 6: Insert shot, 50mm handheld / Rhythmic cut into eggs and toast hitting the pan under warm practical light. / SFX: butter sizzle, chop tap. SHOT 7: MCU, centered 50mm push-in / Match cut into one rushed bite, a quick clock glance, and an immediate rise from the chair. / SFX: crunch, ceramic clink, chair scrape. SHOT 8: Bird's-eye insert, 35mm overhead / Cut on action into striped socks snapping on. / SFX: fabric stretch, heel tap. SHOT 9: MS, 35mm pivot / Camera wipe into a rushed outfit change as the sleep tee disappears under a fitted top and tailored jacket, then her tote, keys, and transit card get scooped up in one messy grab. / SFX: fabric whip, key jingle, zipper pull, bag rustle. SHOT 10: Insert shot, 50mm overhead / Match cut into lace-up shoes slamming on as the laces yank tight in one impatient pull. / SFX: sole thump, lace tug, short breath. SHOT 11: WS, 24mm parallax / Whip pan transition into her, now in the tailored outside outfit, rushing through the apartment door into corridor light without breaking stride. / SFX: latch click, rapid footsteps, hallway air. SHOT 12: MS to CU, 35mm glide into 85mm push-in / Sound bridge into the metro car interior only as she grips the pole, shifts with the carriage sway, checks the passing station lights, and snaps a tense glance toward the closing doors, reflected chrome streaking around her and the city smearing outside the window. / SFX: rail clatter, carriage screech, door warning chime, tight breath. SHOT 13: Insert to MCU, 50mm snap zoom / Smash cut to the office entrance as her access card hits the reader, the glass door unlocks, and she slips through fast before the chair roll and laptop open. / SFX: badge beep, door click, laptop chime. SHOT 14: OTS, 35mm handheld / Rhythmic cut into fingers racing across keys, chat windows blinking, coffee by the trackpad, and notifications stacking faster than she clears them. / SFX: keyboard burst, notification ticks, mouse click. SHOT 15: WS, 50mm pull-out / L-cut with a match from laptop close to apartment re-entry as the jacket drops, work clothes peel away, and she changes back into sleepwear before collapsing into bed in the opening frame shape. / SFX: door shut, bag drop, fabric rustle, blanket rustle, room tone.

Seedance 2.0 AI video prompt: SCENE CONTEXT
A ruined post-war Japanese city plaza. A pink-haired woman and a h

SCENE CONTEXT A ruined post-war Japanese city plaza. A pink-haired woman and a heavy combat mecha start FAR APART: the woman in the near foreground, the mecha 40 meters away in the deep background, a wide empty field of rubble clearly visible between them. The mecha begins launching cluster missiles in an uninterrupted continuous stream — the barrage runs from the first second to the final blow with zero pause. She dodges her way through the relentless missile rain, weaving as she sprints the full 40-meter gap; her right fist swells into a giant fist wreathed in blazing flames, and one flaming punch with devastating impact launches the mecha straight backward, away from the camera, deep into the ruins — and the barrage ends with it. ACTIVE REFERENCES @image1: young woman, athletic build, short pink bob hair, small red horns, black choker, glossy black zip-up catsuit with belt, black gloves and boots. 100% matches the reference. @image2: heavy humanoid combat mecha, dark grey-green segmented armor plating, red glowing vents on chest and shoulders, clawed feet, arms and hands at fixed reference proportions, rendered as a real physical war machine: weathered metal, chipped paint, hydraulic joints. 100% matches the reference. @image3: destroyed Japanese city — collapsed wooden houses and debris fields in the foreground, leaning utility poles, a flooded canal with a damaged stone bridge in the midground, a ruined multi-tiered castle tower and gutted concrete blocks in the background, overcast grey sky. 100% matches the reference. LOCATION MAP One straight axis runs from the camera into the depth of the frame: camera and woman at the near end, the mecha 40 meters down the axis, and behind the mecha a crash corridor of gutted concrete blocks stretching another 100 meters into the background. The ruined castle tower sits far left in the background as a fixed landmark. Light comes from the overcast sky, flat and diffuse. The camera operator stays at ground level on the shadow side and keeps this single axis for the whole scene. FIRST FRAME / BLOCKING Deep-staged wide frame: @image1 stands in the near foreground at frame left, full figure large in frame, back three-quarters to camera, facing down the axis. @image2 stands small in the deep background at frame center-right, 40 meters down the axis, its full body clearly distant — the empty rubble field between them fills the middle of the frame. The near-far relationship reads instantly: she is near, it is far. Handheld frame tilts slightly, horizon imperfect. FORMAT MODE Timed multishot. Cuts only at the specified points, the camera does not cut on its own. Continuous barrage rule: from 0.5s until the punch lands at 11.0s, the mecha's shoulder and back launchers fire cluster missiles in an uninterrupted cycle — new missiles rise every second, missiles detonate on the ground every second. All three states — launch, flight, and explosion — exist simultaneously in every shot. 0.0s to 3.0s — 63° FOV, handheld behind the woman's shoulder, deep-staged: woman large in foreground left, mecha small and distant at center-right, 40 meters of empty rubble between them. The distant mecha's shoulder and back panels snap open; cluster missiles launch upward at 250 km/h in continuous waves, one wave after another with no gap, smoke trails corkscrewing skyward in overlapping layers. As the first missiles detonate around her, new missiles are still rising. She launches into motion immediately, beginning her dodges among the first blast plumes. The 40-meter gap stays visible for the entire shot. 3.0s HARD CUT 3.0s to 6.0s — 84° FOV, handheld tracking the woman. The missile rain pounds down without pause, detonating around her at 10 to 15 meter spacing, the sky always holding the next wave of falling missiles and rising smoke trails. She weaves through the continuous blast plumes at 100 km/h — sliding under one explosion, vaulting a concrete slab, cutting hard sideways to evade two missiles landing at once — every missile striking empty ground. She keeps advancing down the axis toward the distant mecha, which keeps firing in the far background, its shoulder launchers flashing continuously. Successive shockwaves rock the camera. 6.0s HARD CUT 6.0s to 8.0s — 63° FOV, low handheld follow from behind her at knee height, looking down the axis. She sprints straight ahead at 120 km/h while missiles keep detonating ahead, behind, and beside her — dodging as she runs, cutting sharply left and right around blast points at full sprint, mud, tiles and blast debris spraying in her wake. The mecha starts as a small point in the distance and grows steadily in frame as she closes in, its shoulder launchers still spitting missiles and muzzle fire. Real lens flare streaks the frame as the lens crosses the sky's bright patch. 8.0s HARD CUT 8.0s to 9.5s — 29° FOV, handheld close-up on the woman ALONE mid-leap, the background only sky, blurred rubble and blurred continuous explosion flashes, missile shrieks streaking past on all sides. She draws her right fist back and it SWELLS — her glove and forearm inflate in one smooth surge until her fist alone is as large as her entire body, knuckles the size of her head. The instant the swelling completes, the entire giant fist IGNITES: real orange-red flames wrap the stretched glossy surface, fire tongues curling along the knuckles, embers streaming backward, the air around the fist warping with heat. Behind this burning giant fist, her body, face and left arm stay normal size. She hangs one beat at the top of the leap, the burning giant fist cocked, firelight painting her profile in warm tones. 9.5s HARD CUT 9.5s to 12.0s — 84° FOV, handheld low, positioned beside the woman looking down the axis at the mecha. The mecha is still firing — its last wave of missiles has just left the launchers — when her burning giant right fist slams into its chest. On contact a massive fireball erupts with a single deep concussion, flames jetting out along the caving armor seams. The chest armor collapses inward around her knuckles over a wide area, armor plate edges heated to glowing orange-red and melting, droplets of molten metal spraying, a shockwave ring of fire and dust flattening the dust in a 20-meter circle — the impact so violent the mecha's entire torso folds inward for one beat before it launches. The launchers die the instant the torso is destroyed, the barrage ends, and the last few missiles already airborne finish detonating on distant empty ground. The mecha is launched STRAIGHT BACKWARD along the axis, away from the camera, at 200 km/h — trailing fire and black smoke from its chest, shrinking with distance as it flies, arms flailing, holding one straight line down the crash corridor. The camera holds the axis; the mecha recedes deeper into the background for the entire shot. 12.0s HARD CUT 12.0s to 15.0s — 63° FOV, wide handheld on the same axis. Deep in the background, 80 meters away, the mecha crashes down and plows through two gutted concrete blocks, each collapsing in sequence, grinding a long straight trench directly away from the camera before stopping in a rubble mound — flames still burning inside the large hole burned through its chest, black smoke billowing up, red vents dark, sparks and hydraulic fluid spraying and igniting on the open flames, burning debris scattered around the wreck, dust and smoke columns rising along the crash corridor. Only dissipating missile trails remain in the sky. In the near foreground the woman lands in a crouch; the flames on her right fist die down to embers, the fist deflates back to normal size in one smooth breath, thin smoke and steam rising off the glove, and she slowly stands with her back to camera, looking at the distant burning wreck. Dust density climbs from 20% to 60%. CAMERA Documentary operator, handheld throughout: constant 1–3cm shake, imperfect reframes, hunting focus, half-beat lag behind the action. The continuous explosions rock the camera repeatedly in every segment. One spatial axis for the whole scene — every shot looks down the same line from woman toward mecha. No drift mid-segment; FOV locked per segment. Camera-native tonal response: flat natural contrast, highlights clip honestly on explosions and flames, deep shadows stay soft. PHYSICS The woman moves with real mass — feet dig into mud, debris scatters, inertia in her hair and the catsuit's gloss. Missiles fall in continuous waves along gravity arcs, adjacent blast shockwaves stacking, pushing loose tiles outward layer by layer. Her burning giant fist carries visible mass and real fire behavior: flame tongues stretch backward in the airflow as it swings, embers scatter along its motion path, heat warps the air around it. At impact: armor collapses along edge-stress fractures, heated metal glows orange-red and drips, the fireball expands as a real deflagration then rolls into black smoke. The launched mecha follows one straight ballistic line backward along the axis — rising, peaking, falling — losing speed through each concrete block it destroys, collapses running edge-to-center, sequential. Ground tremor shakes the camera on the final crash. LIGHTING Single source: overcast sky, flat diffuse daylight, 5600K fixed. Natural unpolished exposure — grey sky slightly blown, faces lit softly from above. The continuous missile detonations provide constantly flickering warm pulse light across the whole scene; the burning giant fist, the fireball and the burning wreck act as intense warm practical light sources, spilling dancing orange firelight onto the wet mud, her glossy catsuit and the mecha's dark grey-green armor, with real fire reflections on the glossy materials. Real lens flares when the lens crosses the sky's bright patch or the fireball. AUDIO Wind over rubble, distant creaking metal. The barrage never stops: rapid pneumatic launch thumps and overlapping ground detonations stack in continuous layers, missile flight shrieks streaking past throughout — until the punch lands at 11.0s and it all stops abruptly, leaving only a few final distant blasts. Her sprint: fast footfalls and breathing threaded between the explosions. The fist swelling and igniting: a taut stretching creak, then the whoosh of flame ignition and a sustained fire roar. The punch: one massive bass slam layered with a deflagration boom, tearing metal, then the whistle of a flaming mass receding into the distance, two sequential building collapses far away, and a long grinding skid, with residual fire crackling in the tail. Handheld mic character — clipped peaks on explosions. STYLE Realistic documentary footage, true-to-life untouched colors straight from camera, natural unpolished lighting, subtle motion blur on fast movement, photoreal texture on skin, latex, mud, fire and weathered metal. News-footage authenticity. POSITIVE LOCKS Opening geometry: the woman is near the camera and the mecha is 40 meters away in the deep background, both visible in the same frame with empty rubble between them. Barrage continuity: from 0.5s to 11.0s the mecha fires missiles continuously in every shot — missiles in flight and smoke trails in the sky in every frame, new explosions on the ground every second; launching ends when the punch lands at 11.0s. The giant fist belongs to the woman only — it is her right fist; flames appear only on her right fist and forearm, from 8.0s to 12.0s, and after the punch only on the mecha's chest and the burning wreck. The mecha's arms and hands stay at their fixed reference proportions in every shot; the mecha's arms stay at its sides or flailing — its only actions are launching missiles and being sent flying. Her body, face and left arm stay normal size and unharmed at all times. After the punch the mecha travels in ONE straight line backward along the axis, away from the camera, trailing fire from its chest, shrinking with distance in every frame, and stays down after the crash, still burning. @image1's catsuit, pink hair and red horns identical in every shot. @image2's armor and red vents match the reference until the torso is damaged. Every missile lands on empty ground; the woman stays untouched. One overcast time-of-day and weather throughout. Single camera axis for the whole scene. Handheld shake in every segment.

Seedance 2.0 AI video prompt: SUBJECT: image1
WARDROBE: Sleep tee and lounge shorts at home. Tailored jacket,

SUBJECT: image1 WARDROBE: Sleep tee and lounge shorts at home. Tailored jacket, trousers, and lace-up shoes outside. ENVIRONMENT: Tiny apartment → fridge glow rain-dusted hallway chrome metro clean office bedroom (cool window light). Glossy, lived-in. MOOD: Late-for-work panic → fast momentum → breathless urgency exhausted exhale MUSIC: Fast percussive electro-pop COLOR LOGIC: Hyperreal Pop Look STYLE: Ultra-Realistic LOGIC RULE: Maintain continuity in wardrobe, props, locations, and action SHOT 1 - ECU, 85mm push-in 06:50 on phone screen shaking on rumpled sheets SFX: alarm, sheet rustle SHOT 2-WS, 35mm handheld He jolts upright, throws blanket aside, feet hit floor in one motion (sleep tee + shorts) SFX: mattress bounce, blanket whip, sharp breath SHOT 3-MCU, 50mm slide Face wash at sink, water droplets catching light SFX: faucet rush, water slap SHOT 4 - Insert, 85mm rack focus Toothbrush, mint foam, mirror eye SFX: bristle scrape, sink drip SHOT 5 Fridge POV, 24mm Door snaps open, his hand grabs ingredients in blue light SFX: fridge hum, bottle clink SHOT 6 - Insert, 50mm handheld Eggs + toast hit pan SFX: butter sizzle SHOT 7 - MCU, 50mm push Quick bite clock glance stands ← immediately SFX: crunch, chair scrape SHOT 8 - Overhead, 35mm Socks snap on SFX: fabric stretch SHOT 9-MS, 35mm pivot Fast outfit change tailored jacket on grabs bag, keys, transit card SFX: fabric whip, key jingle SHOT 10 Insert, 50mm Shoes slam on, laces pulled tight SFX: thump, lace tug SHOT 11-WS, 24mm parallax Rushes out into hallway without breaking stride SFX: latch click, footsteps SHOT 12-MS CU, 35-85mm Metro: gripping pole, shifting with motion, tense glance at closing doors, chrome reflections SFX: rail clatter, door chime SHOT 13- Insert MCU, 50mm Office entry: card scan → door unlock → slips in laptop opens SFX: badge beep, click SHOT 14-OTS, 35mm handheld Typing fast, notifications stacking, coffee beside trackpad SFX: keyboard burst, ticks SHOT 15- WS, 50mm pull-out Back home jacket drops changes to sleepwear collapses into bed (same opening frame) SFX: door shut, bag drop,

Seedance 2.0 AI video prompt: First-person action-camera POV, wide-angle lens with mild distortion — NOT a fis

First-person action-camera POV, wide-angle lens with mild distortion — NOT a fisheye: no black or vignetted corners — camera mounted at the wearer's chest: very shaky handheld throughout — constant hard jitter and bounce, the frame bobbing and swaying violently with every step, whipping hard on impacts and sprints, never smooth, never stabilized, never tripod-still. The wearer is a medieval peasant man; his weathered hands with dirt-caked nails and coarse brown wool sleeves enter the frame constantly and stay consistent in every shot. Cold medieval England, early-morning soft light, wet deep mud everywhere with sheen and ruts, chimney smoke, thatched and timber-framed village below a massive grey stone castle on a hill. Mud behaves physically: splatter arcs, heavy clinging clumps, wet squelch. Raw unedited camera footage, natural light only, no color grading, not cinematic. No on-screen text, no subtitles. No background music, diegetic sound only. No cuts within a shot. the woman is @ Shot: First-person action-camera POV, wide-angle lens with mild distortion (no fisheye, no black corners). Waking at dawn in a timber barn loft, thin blades of light through gaps in the plank walls, straw and grey fur. The wearer's head turns and his young wife is right there beside and close him — early 20s, loose brown braid, freckled skin — just waking herself: she rubs one eye with the back of her hand, breaks into a wide sleepy yawn, stretches a little, blinks slowly and settles into a soft drowsy smile. She murmurs, in a weak, soft, just-woken sleepy voice, half a yawn still in it, looking into the lens (on screen, lip-synced): "Stay a while... the day can wait." From behind the camera he answers, warm and low, off-screen POV (no lip-sync): "The lord's fields can't, love." She gives a lazy grin and hooks a hand behind his neck to pull him closer, his weathered hand rising into frame to her cheek, and she teases, still sleepy and soft: "Then let the lord plough his own field." He leans in and kisses her, the frame dipping in close, her hand sliding to the back of his head, a soft shared laugh. Warm, unhurried, natural soft daylight. No cuts within a shot. No background music. First-person action-camera POV, wide-angle lens with mild distortion — NOT a fisheye: no black or vignetted corners — camera mounted at the wearer's chest: very shaky handheld throughout, constant hard jitter and bounce, the frame bobbing and swaying violently with every step, whipping hard on impacts and sprints, never smooth, never stabilized. The wearer is a medieval peasant man; his weathered hands with dirt-caked nails and coarse brown wool sleeves enter the frame constantly and stay consistent. Medieval England, cold damp climate, wet deep mud with sheen and ruts, chimney smoke, thatched and timber-framed village below a massive grey stone castle on a hill; natural light only. Mud behaves physically: splatter arcs, heavy clinging clumps, wet squelch. Raw unedited camera footage, no color grading, not cinematic. No subtitles, no on-screen text, no background music (diegetic sound only). No cuts within a shot. Shot 1: Cold early morning. Looking down at a deep-rutted mud road while walking, boots squelching, wet mud glistening; the wearer's weathered right hand holds a gnarled walking stick and a small tied burlap sack. The camera tilts up mid-stride revealing a massive grey stone castle with flags on a hill above a medieval village, horse carts and hooded villagers on the road ahead; a passing villager brushes close past the lens. Frame bobs with every step. Shot 2: Same POV arrives at a muddy market square. A bald man with a thick reddish-grey beard in knee-length chainmail and a grey cloak stands holding a tall staff beside a wooden cart piled with bulging burlap grain sacks, a tray of eggs and live brown chickens; a clerk's hand writes on parchment at the right edge of frame; crowd murmurs. The wearer's weathered hands reach into frame and cinch a burlap sack shut on the cart edge. Shot 3: Same POV. An old woman in a dark grey hooded shawl, 70s, deeply lined face, bushy grey brows, shrewd eager eyes, hurries up close through the mud; she points a bony finger down at the eggs and looks up at the lens, asking in a cracked voice (on screen, lip-synced): "How much for five?" From behind the camera the wearer answers plainly, off-screen POV (no lip-sync, face not seen): "A penny." She digs a small coin from a cloth purse and presses it into the wearer's open weathered palm as his hand comes into frame to receive it. Behind her hang furs on a stall, clay pots, a pig rooting in the mud.

Seedance 2.0 AI video prompt: Image1 is the main character — maintain consistent facial features and body type

Image1 is the main character — maintain consistent facial features and body type throughout. Cinematic time-freeze short film, 15 seconds, ultra-realistic, Arri Alexa Mini shooting texture, 50mm lens, rain-slicked evening streets with amber streetlight and neon reflections, shallow depth of field. [0:00–0:03] Rainy evening. A crowded pedestrian crossing in the city center. Normal time flow. Steadicam front-facing medium shot tracking: the main character in a dark charcoal overcoat walks through the crowd without an umbrella. Rain falls straight down. People hold umbrellas, rush across the road, stare at phones. A taxi cuts through a puddle in the background, water spraying up. He reaches the center of the crossing, raises his right hand — and snaps his fingers. [0:03–0:06] The instant of the snap — a white spherical shockwave bursts from his fingertips, carrying visible air distortion and light refraction, expanding outward in all directions. Every raindrop stops in place. Thousands of suspended water drops hang motionless at different heights, catching the amber streetlights and neon signs, turning the air into a dense field of tiny glowing beads. The taxi’s splash locks mid-arc above the pavement. Umbrella fabric stops mid-sway. Puddle ripples freeze flat. After the shockwave disappears — total silence. The suspended raindrops shimmer in the warm streetlight. [0:06–0:09] Only his footsteps sound against the wet pavement. Camera pulls backward ahead of him — he walks forward through the frozen crossing, moving between suspended pedestrians and frozen umbrella spokes. He reaches out and touches a hanging raindrop with one fingertip. The drop holds its position, perfectly spherical, reflecting a miniature upside-down street inside it. He leans in and looks at his own reflection in the surface. Straightens up, puts his hand back in his pocket, keeps walking. [0:09–0:11] He stops in front of a street vendor’s cart. The vendor is frozen mid-pour — a thin stream of hot liquid suspended in the air above a paper cup, steam curling up in a still ribbon. He lifts the cup from the cart and slides it under the frozen stream. Steps back. Tilts his head. Studies it. Quietly says: “Almost.” [0:11–0:15] He sets the cup down, turns to face the camera directly, raises his right hand — and snaps again. A second shockwave, stronger, bursts outward in reverse. All frozen rain resumes at once — a sudden rush of falling water hits the pavement in a single loud pour. The splash unfreezes and slaps the ground. The vendor finishes his pour, blinks at the cup, finds it in the wrong spot. Umbrellas swing. Pedestrians keep moving without a pause. City rain sounds return. He turns and walks into the rain. Camera rises into an overhead wide shot — his figure moves through the crossing and disappears into the crowd. Fade to black. Sound design: Rain and city traffic — finger snap triggers a low-frequency shockwave rumble radiating outward — total silence — wet footstep echoes — single raindrop tap — whispered “almost” — second finger snap — stronger reverse shockwave explosion — rainfall resumes all at once in a rush — city sounds fully restored

Seedance 2.0 AI video prompt: SHOT 01 Ultra-low front tracking shot inches above asphalt, a high-performance s

SHOT 01 Ultra-low front tracking shot inches above asphalt, a high-performance sports car already at full speed blasting through a tight urban street, suspension compressed, reflections streaking across glossy body panels, extreme motion blur on surroundings, no buildup, immediate velocity, hyper-realistic, cinematic, 8k detail. SHOT 02 Side close tracking shot as the car threads between parked vehicles at high speed, speed maintained, tight urban environment, photorealistic, intense motion. SHOT 03 First-person FPV drone aggressively chasing directly behind the car, locked onto rear bumper, rapid swerves through traffic, slight camera instability, extreme speed sensation, kinetic, photorealistic. SHOT 04 Extreme close-up on front wheel during hard braking into a corner, ABS pulsing rapidly, tire briefly losing and regaining grip, subtle instability, rubber deformation visible, asphalt texture sharp, hyper-detailed, realistic physics. SHOT 05 Front 3/4 angle as the car exits the turn, driver overcorrects steering, rear stepping out wider than intended, uneven weight transfer, tires struggling for grip, tension rising, near loss of control. SHOT 06 Rear tracking shot as the car transitions into a deep uncontrolled slide, rotation increasing, tires fully slipping, appearing on the edge of spinning out, smoke beginning to form, maximum instability. SHOT 07 Same motion continued, but the driver countersteers precisely at the last moment, front wheels snapping into alignment, grip suddenly returning, the slide stabilizing just before full spin, razor-thin recovery, physically realistic. SHOT 08 Side tracking shot as the car shoots out of the drift with regained traction, tires gripping hard, smoke trailing behind, acceleration surging forward, control re-established under extreme force. SHOT 09 Ultra-low tracking shot again, even faster now, the car blasting past the camera with inches to spare, air displacement and motion blur overwhelming the frame, pure speed dominance. SHOT 10 Wide shot of the street ahead as the car disappears into the distance, leaving faint tire smoke and echoes of motion, environment settling, tension released without impact.

Style: Ultra-realistic mass celebrity arrival scene. Single continuous shot. Handheld camera from crowd perspective. Natural micro-shake. No cuts. Documentary-level realism. Audio: Only natural environment sound loud crowd cheering, overlapping voices shouting, rapid camera shutter clicks, phones recording audio, distant city noises, footsteps, fabric movement. Lighting: Nightime, illuminated from street lights. Mixed reflections on polished surfaces. Soft but realistic shadows. Slight atmospheric haze for depth. 

Main Character: (Subject Reference Image) Calm, controlled presence. Subtle confident smile. Face identity must remain perfectly consistent across all frames. Outfit: Exactly the same from reference image. Location: An event in New York City.

Scene Flow 

0–3s: Camera starts from inside a dense crowd behind barricades. Handheld, slightly unstable. View partially blocked by people in front. Multiple phones raised, some screens visible recording. Crowd energy is loud, chaotic, restless.

3–6s: Camera lifts slightly above shoulder level, still handheld. Focus shifts naturally between heads, waving hands, and we see our subject exit a fancy hotel in New York City. Occasional bright flashes from media cameras. Anticipation rises as crowd leans forward.

6–10s: Security personnel step in, pushing the crowd back slightly. Camera reacts with natural shake. Through shifting gaps, the main character appears in the distance—initially soft and partially obscured, gradually becoming clearer while walking forward with a small escort team. 

He walks up to a fan and signs printed photo she is holding of (Subject Reference Image) 10–13s: Subtle handheld push-in (natural movement, not digital zoom). The main character is now clearly visible, walking confidently at center frame. Path is being cleared. He raises one hand and gives a calm, controlled wave with a slight smile. Camera struggles slightly to keep framing due to crowd movement.

13–15s: Camera shifts and tilts trying to follow. A luxury convoy becomes partially visible featuring three large premium SUVs. A security member opens the back door of a the suburban in the middle of the three. The main character enters quickly and rolls the window down. Vehicles begin moving forward as he waves to the crowd. Camera lifts slightly as people jump and try to capture the moment.

Seedance 2.0 AI video prompt: Create a 15-second nostalgic cinematic montage based on the storyboard sequence.

Create a 15-second nostalgic cinematic montage based on the storyboard sequence. A quiet young Japanese woman spends one calm day alone in 1990s Tokyo. Keep the same girl consistent throughout all shots: long dark hair with bangs, wearing a loose white dress, carrying a light canvas tote bag, with a soft and thoughtful expression. Follow this sequence in quick, elegant cuts: 1) Morning train interior — medium shot of her sitting by the window reading a paperback book as soft sunlight moves across the carriage. 2) Old bookstore district — rear tracking shot of her walking down a narrow Tokyo street lined with secondhand bookstores, faded signs, power lines, and stacked books. 3) Inside a used bookstore — intimate side shot of her browsing shelves and opening a book she finds. 4) Outdoor stairway — low angle shot from behind as she slowly walks up a narrow residential staircase lined with plants and old houses. 5) Convex mirror shot — show her reflected in a roadside traffic mirror, pausing and quietly looking upward. 6) Retro kissaten café interior — wide shot showing the full atmosphere of the café: warm lights, vintage booths, framed decorations, window light, and the girl sitting alone at a table. 7) Cream soda close-up — close-up of her softly sipping a bright green cream soda with vanilla ice cream and a cherry. 8) High viewpoint over Tokyo — wide emotional shot of her standing at a railing, looking over train tracks and the Tokyo skyline at sunset. 9) Night train home — quiet final shot of her riding the train at night, softly lit, reading again, with her reflection visible in the window. Style: realistic, cinematic, nostalgic 1990s Tokyo, soft natural light, subtle film grain, slightly faded colors, calm poetic mood, everyday realism, gentle emotional tone. Camera style: smooth handheld and slow tracking shots, subtle push-ins, natural motion, varied framing, elegant pacing, realistic movement, no exaggerated acting. Editing rhythm: fast but emotional montage, each shot brief and memorable, flowing naturally from morning to evening.

More use cases

AI Video Prompts for AdsAI Video Prompts for AnimeAI Video Prompts for Product ShotsAI Video Prompts for Aerial & Drone ShotsAI Video Prompts for Action ScenesAI Video Prompts for Nature & LandscapeAI Video Prompts for FantasyAI Video Prompts for HorrorAI Video Prompts for Documentary StyleAI Video Prompts for Romance & Love StoryAI Video Prompts for Surreal & Dreamlike VideoAI Video Prompts for Sci-FiAI Video Prompts for PortraitAI Video Prompts for Food & CookingAI Video Prompts for Real Estate & ArchitectureAI Video Prompts for FashionAI Animation & 3D PromptsAI Video Prompts for Animal VideoAI Video Prompts for DramaAI Video Prompts for Wedding & CelebrationCinematic AI Video PromptsAI Video Prompts for TimelapseAI Video Prompts for Tracking ShotsAI Video Prompts for DanceAI Video Prompts for ComedyAI Video Prompts for Slow MotionAI Video Prompts for First-Person & POVAI Video Prompts for Retro & Vintage VideoAI Video Prompts for Close-Up VideoMinimalist AI Video PromptsAI Video Prompts for Travel VideoAI Video Prompts for Abstract VideoAI Video Prompts for Zoom ShotsAI Video Prompts for Pan ShotsAI Video Prompts for Urban VideoAI Video Prompts for Fitness & WorkoutAI Video Prompts for Clay Animation & ClaymationAI Video Prompts for Night ScenesAI Video Prompts for Transformation & MorphingAI Video Prompts for SportsAI Video Prompts for Underwater ScenesAI Video Prompts for Music VideoAI Video Prompts for Rain & Wet WeatherAI Video Prompts for Fog & MistAI Video Prompts for Fire & FlameAI Video Prompts for Car & Automotive VideoAI Video Prompts for Golden HourAI Video Prompts for Snow & Winter VideoAI Video Prompts for Space & Cosmic VideoAI Video Prompts for Creature & Monster VideoAI Video Prompts for VFX & Time Freeze EffectsAI Video Prompts for Landscape VideoAll categories

Frequently asked questions

What are the best AI video prompts for realistic video?

The best realistic video prompts name the camera hardware by brand and model ("mini DV camcorder," "iPhone 14 Pro front/back camera mix," "GoPro Hero chest mount") rather than a generic "handheld" style. The hardware spec activates the full artifact bundle — focus hunting, tape grain, rolling shutter, autofocus breathing, HDR behavior — simultaneously. Pair the hardware spec with an exclusion list ("no cuts, no color grading, no beauty filters, no on-screen text") to override AI defaults that read as synthetic. Every prompt in this gallery combines both layers.

How do I make AI video look realistic with Seedance 2.0?

Two techniques together produce the most reliable realism. First, name the recording hardware ("mini DV," "iPhone 14 Pro," "doorbell camera") rather than describing a visual style — the hardware name activates an entire system of physical artifacts that aesthetic descriptions like "handheld" do not. Second, write an exclusion list: "no cinematic color grading, no beauty filters, no skin smoothing, no third-person shots, no artificial HDR, no on-screen subtitles." AI video defaults to "improved" footage; the exclusion list overrides each default behavior individually. For scenes with physical materials — water, sand, cloth, hair — add explicit physics specs: "realistic splash physics, no exaggeration," "hair responds dynamically to wind," "sand particles scatter naturally at footfall."

Can Seedance 2.0 generate video that looks indistinguishable from real footage?

Yes. Seedance 2.0 can generate footage that reads as real when the prompt names the camera system, overrides the AI defaults via exclusion lists, and specifies the physical simulation behavior of materials in the scene. The key shift is treating realism as a camera direction problem, not a resolution problem: "photorealistic" as a standalone adjective produces clean AI-look output; "iPhone 14 Pro, front-camera selfie footage mixed with rear-camera event coverage, natural handheld shake, micro-shakes, autofocus breathing, rolling shutter wobble, no color grading" produces footage with the specific physical signature of a real smartphone camera. Browse the realistic video prompts here for examples with preview videos.