YouMind
Sign in

Documentary

Explore free Documentary video prompts from YouMind's AI prompt library, grouped under styles. Every prompt is curated from real creative workflows and ready to copy, adapt, and reuse.

745 promptsDaily highlights
Seedance 2.0
View full prompt & copy

15-second ultra-realistic funny short video, 9:16 vertical screen, mobile phone original camera documentary texture, real gym environment, soft indoor top light, the whole process is like a friend casually following along, slight natural shaking, very strong daily life feel. [Character Setting] High-looking young Asian woman, natural light makeup, delicate and relaxed facial features, well-proportioned figure, gentle and cool temperament, real passerby feel, skin retains natural texture and subtle sweat beads, no influencer smoothing, no exaggerated plastic surgery face, natural body proportions. Face and body completely consistent with reference image. Shot 1 | 0-4s Fake Self-Discipline Bright and clean gym, background has few training figures and equipment. Girl doing standard squats seriously in free training area, movements smooth and stable. Camera handheld tracking from right rear, first slowly retreating to keep full body in frame, then slightly lifting as she stands up, like a friend walking and filming, picture has slight breathing feel and natural shaking, creating illusion of her being very self-disciplined and hardworking. Shot 2 | 4-8s Misleading Continued Training Hard cut to side tracking shot, girl standing holding small dumbbells doing bicep curls, slightly sweating, serious exertion, calm and focused expression. Camera close to her side front, first slight lateral movement, then slowly pushing close to upper body, background equipment naturally passing by, audience thinks this is a self-disciplined fitness video. Shot 3 | 8-12s First Layer Twist Girl suddenly puts down dumbbells, expression changes from focus to relaxation, long sigh. Camera continues handheld following her to training bench. She sits down, takes out fried chicken and iced milk tea from bag, actions natural and skilled, tears open packaging. Camera quickly but slightly pushes close when she takes things out, forming candid comedic effect. Shot 4 | 12-15s Ultimate Twist Camera directly pushes close to face and fried chicken/milk tea in same frame. Girl naturally bites fried chicken, drinks milk tea, looks at camera, says in casual tone: "Eat full to have strength to lose weight." Expression calm afterwards, like saying something normal. Ending stays on her calm eating contrast face, handheld camera has extremely slight forward/backward shaking, enhancing friend's secret filming feel. [Visual Quality] Ultra-realistic live-action, original camera HD quality, real skin texture, pores, sweat beads and natural skin tone, slight environmental noise in gym, natural indoor top light, do not use movie-grade refined lighting. Camera movement mainly handheld tracking, slight push-in, slight lateral movement, rhythm casual, no gimbal smooth motion. Direct hard cuts between scenes, no fancy transitions. [Sound Effects] Gym distant equipment collision sounds, slight footsteps, girl's training breathing sounds, packaging tearing sound, fried chicken chewing sound, milk tea straw sound. Dialogue clear and natural, no exaggerated dubbing feel. Add short light funny music point after dialogue ends. [Negative Prompts] Deformed fingers, twisted limbs, extra fingers, missing fingers, face collapse, misaligned facial features, AI fake face, plastic skin, excessive beauty filter, silicone face, exaggerated contact lenses, deformed body, disproportionate ratio, abnormal head-to-body ratio, floating objects, object clipping, screen flickering, object drifting, cartoon rendering, 3D animation, hand-drawn, illustration, screen subtitles, text, fancy fonts, watermarks, logos, borders, strong filters, soft light smoothing, stabilizer smooth motion, fixed camera position, stiff posing, stage lighting, pale fake light, excessive lighting, extra people, messy distracting clutter, exaggerated performance, awkward acting, affected expressions

Generate video
Seedance 2.5
View full prompt & copy

Small depot in mountain valley. White steam creeps from under black steam locomotive. Sakura looks up from cab window. Old man shovels coal into firebox. Pulling lever sharply causes driving wheels to spin idle, smoke erupts explosively. Cab becomes pure white. As steam clears, Sakura coughs with soot-covered face. Old man laughs, gently places gloved hand over hers, shows how to return lever slightly. With sooty face, opens lever slowly. Chimney puffs, driving wheels grip rails and start turning. Leaning out window, hair flows in wind. Pulls whistle cord, white steam shoots straight up. Small white flag near tracks approaches. High-five in cab. Laughing with sooty faces. Record shot handheld by friend who came along. Approx. 2 seconds per cut, no dialogue/subtitles/BGM. Only two people appear on screen. One locomotive, one lever, one whistle cord. Number plates/body/clothes are plain with no text. Shape of locomotive, yard layout, cab interior fixed by reference image without people.

Generate video
Seedance 2.5
View full prompt & copy

Autumn rice field. The cutting part of a combine harvester approaches, pushing through the rice plants. Sakura in the driver's seat widens her eyes. An old man on the ridge extends his arm straight forward. When he pushes the lever, the machine veers away, then veers too close. Looking down from the embankment, the harvested track winds like a snake. The old man crouches and traces the rows of rice stalks with his palm. Re-ties chin strap, grips the lever. Aligning the tip of the divider plate with the row of stalks, the machine advances straight without shaking. The last rice plants are sucked in. Looking down, next to the winding track is a single straight harvested line. Golden straw flows like a waterfall onto the bed of the light truck. Raises both hands in the driver's seat. Record shot handheld by a friend who came along. Approx. 2 seconds per cut, no dialogue/subtitles/BGM. Only two people appear on screen. One combine harvester, one light truck. Machine/clothes/hat are plain with no text. Shape of combine and layout of rice field fixed by reference image without people.

Generate video
Seedance 2.0
View full prompt & copy

Vertical 9:16 hyper-realistic transformation video, documentary style, static wide-angle camera from the corner of an empty white room with one bright window and a bare concrete floor. Shot 1 (0–4s): A bearded man in blue denim overalls and a black long-sleeve shirt rakes a huge cone-shaped pile of thousands of antique brass and iron keys with a black leaf rake, methodically spreading the keys evenly until the entire floor is covered in a dense carpet of old keys. Shot 2 (4–8s): A woman with long brown hair in red overalls and black gloves pours thick glossy white self-leveling epoxy resin from a large white plastic bucket; the milky liquid flows outward in smooth waves, swallowing the keys and covering the whole floor with a mirror-smooth white surface, a few key silhouettes poking through. Shot 3 (8–17s): A worker in a yellow high-visibility vest, black cap and grey work pants pushes a red-and-white ride-on floor grinder with a black vacuum hose, polishing the hardened epoxy in smooth passes; golden key shapes emerge embedded in the glossy floor like terrazzo, reflections of the window and the worker visible on the mirror-like surface. Shot 4 (17–23s): Two workers in black polo shirts and black caps assemble a dark walnut bed frame in the center of the room, place a white mattress, smooth out white bedding and pillows, then wooden nightstands with white ceramic lamps appear on both sides; the finished bed with beige accent pillows stands on the polished floor full of golden keys as the workers walk out of frame. Bright natural daylight, soft shadows, satisfying ASMR renovation timelapse mood, photorealistic textures, smooth motion, no text, no watermark.

Generate video
Seedance 2.5
View full prompt & copy

An autumn sunny day at a heavy machinery experience park. A giant bucket stabs into the ground, dirt flying. Sakura's eyes widen in the driver's seat. The instructor gestures 'slowly' with his palm. The arm extends jerkily. The claw passes through the hole in a tire and stabs into a pile of dirt. The pile collapses, pushing the tire away. Sakura freezes, the instructor laughs. As the dust settles, the tire is right next to the stake. She tightens her chin strap and grips the lever. This time the arm descends smoothly. The claw catches on the tire hole. It lifts up, hanging the black tire in the air. Slowly moves left, directly above the stake. Lowers slowly, the stake goes through the hole, thud. She raises both hands in the driver's seat. Recorded casually by a friend who came along. About 2 seconds per cut, no dialogue, subtitles, or BGM. Only two people appear on screen. One excavator, one stake, one tire, one pile of dirt. Machine, helmet, and clothes are plain without text. Fix the layout of the experience park with a reference image of the wide shot without people.

Generate video
Seedance 2.5
View full prompt & copy

Japanese traditional haunted house during Halloween. In a hallway lined with pumpkin lanterns, a ghost jumps out from a closet raising both sleeves. A couple chuckles and passes by. The ghost drops shoulders back into the closet. Lifting the curtain reveals Sakura's face. Lips sealed tight. Now waiting too long, raises sleeves behind two passing housewives. They chat and leave. Senior staff member taps shoulder, flashes flashlight twice 'This is the signal'. Inside the closet, steadying breath and nodding. From the gap, two boys with devil horns and capes approach fearfully. Behind shoji screen, senior's light goes blink-blink. The two jump up screaming and hug each other. Running away. The remaining ghost is startled by screams and falls butt-first. Curtain flips up revealing wide-eyed face. Grabbed by senior's hand to stand up, shy smile. Returns to closet. Recorded handheld by staff friends. Each cut approx 2 seconds, no dialogue/subtitles/BGM. Only one ghost (Sakura). Closet on left, senior's shoji on right, guests always move front-to-back in one direction. Costume matches reference image draped over faceless mannequin.

Generate video
Seedance 2.0
View full prompt & copy

[Video Specifications and Overall Style] 9:16 vertical screen, 15 seconds, single continuous shot. A thrilling scene accidentally captured by a tourist with an ordinary phone inside a Chinese wildlife park. Initially calm, the photographer reacts half a beat late to a giant crocodile's ambush, causing camera shake, misalignment, and hurried re-framing of the subject. Retain the authenticity of amateur mobile footage. [Character and Reference Locking] The female protagonist uses preset adult Chinese female character #1 @598c0e35-7c27-4b7c-8bac-19bb206ff1c3, wearing preset outfit #2 @d57ec59a-c0a5-4eae-84f4-4f02ef21a92d, strictly maintaining the facial features, long hair, clothing style, color, and material from the references. #1 is the person, #2 is the outfit; numbers do not appear in the frame. The administrator is a Chinese male around forty years old, short black hair, wearing khaki short-sleeved work clothes, dark trousers, with a work badge on his chest, standing behind and to the side of the female protagonist. There are a few adult Chinese tourists in the background, wearing summer casual wear, naturally scattered along the boardwalk. [Scene and Camera Position] Tropical crocodile viewing area, old wooden boardwalk extending along a deep green pool, opposite bank lined with bamboo, palms, and shrubs. Thick wooden railing extends diagonally from bottom right to background. External photographer stands about one meter behind and to the side of the female protagonist, slightly higher than her hand-held phone position, shooting vertically downwards. The right side of the frame shows the female protagonist, administrator, and boardwalk; left and bottom show the pool. Humid daytime, natural sunlight filtering through vegetation, slight green reflection on water surface. Ordinary phone 1x main camera, slight hand shake, tilt, compression artifacts, noise in shadows, brief blur during rapid movement. No cinematic grading, drone shots, or portrait bokeh. [Female Protagonist's Phone Orientation - Must Be Accurate] The female protagonist is using the rear camera to film the crocodile in the water, not taking a selfie, nor showing the phone to the external photographer. She holds the phone horizontally with both hands at the ends, fingers avoiding the rear camera. The phone is below her chest, near the outer edge of the railing, clearly tilted downwards, with the rear lens direction accurately aimed at the huge shadow underwater in the lower left. Phone screen faces up, slightly towards her face; back and rear lens face down towards water. She looks down at the screen, observing the target in water through it. From the external photographer's slightly higher side-rear angle, mainly see the phone screen and side edges, should not clearly see the entire phone back and camera module facing the external photographer. Screen content matches real water surface: same water patch and blurry shadow visible, no selfie face, irrelevant scenery, or wrong direction. When adjusting wrists slightly, rear lens always follows underwater shadow. Only when crocodile attacks does she pull back phone in fear, disrupting orientation. [Crocodile and First Frame State] Only one abnormally large dark gray-black crocodile. When leaping, head size approaches upper body of an adult, open mouth occupies nearly half left frame, size reference original video, no small crocodile generated. First frame: crocodile fully submerged, only huge blurry long shadow like sunken wood. Water reflections and turbidity hide details, cannot clearly identify head, eyes, nostrils, scales, teeth, nothing exposed above water. Crocodile maintains same volume throughout, no sudden expansion during leap. [0-2s: Filming Underwater Shadow] Female protagonist stands inside railing, leaning forward slightly, holding phone horizontally, aiming rear lens down at huge shadow in lower left water. She looks down at screen facing her, shadow in screen corresponds to shadow in pool. External photographer captures phone screen, her profile, and water below from side-rear. Administrator seriously warns: "Ma'am, step back, don't lean on the railing!" Water has only slight ripples, crocodile still hidden. [2-4s: Target Approaches, Phone Follows] Huge underwater outline slowly approaches female protagonist's position. She adjusts wrist slightly, keeping rear lens aimed at moving shadow, screen facing her, not turning phone back to photographer. Administrator steps closer, reaching out to support her, tone urgent: "That's not wood, it's alive! Don't lean in!" She starts pulling body back but phone still filming water. End of segment: crocodile snout and eyes just breaking water surface. [4-6s: Giant Crocodile Ambush] Crocodile violently bursts water surface, huge head rises along outside of railing, open mouth instantly occupies nearly half left frame, splashing large amount of water. Female protagonist screams, hurriedly pulls back phone and arm. Phone orientation loses stability due to shock, must have been correctly aimed before. Crocodile briefly bites into her forearm extended outside railing and nearby fabric. Administrator immediately supports her upper arm and shoulder/back, stabilizing her center of gravity inside railing. External photographer retreats half step in shock, camera briefly shakes upward, no cuts or wound close-ups. [6-9s: Escape and Fall] Crocodile briefly releases grip, head drops back, female protagonist seizes chance to retract injured forearm, administrator pulls her to inner boardwalk. Action must have clear release, arm cannot pass through closed teeth. Phone slips from hand, flips once, falls into pool. Crocodile slams back into water, white splash. Female protagonist covers forearm, small amount of red liquid seeps from injury, forming short thin streaks, staining fabric at contact point, one or two drops falling onto wooden boardwalk. No deep wounds shown, no spraying, no heavy bleeding. Administrator shouts: "Step back! Don't reach out! Don't pick up phone!" [9-11s: Tourist Panic] Female protagonist retreats to middle of boardwalk, protecting injured forearm, shoulders tense, breathing rapid. Nearby tourists first shrink back, grab companions; those further away turn upon hearing scream, some put away phones, others make way, exclaiming sequentially: "Step back quickly!" "Don't lean on railing!" Crowd reaction has time lag, not synchronized, no clipping, no falling into pool. Crocodile remains underwater, no second attack. [11-13s: Administrator Evacuation] Administrator stands between female protagonist and railing, continuing to protect her retreat, gesturing for tourists to make way: "Back up! Everyone don't crowd!" A nearby tourist offers tissue, female protagonist accepts and presses same injury spot, tissue gradually stains red. Photographer retreats slightly and corrects composition, capturing female protagonist, administrator, and panicking retreating tourists. [13-15s: Aftermath] Female protagonist covers forearm, gasping, trembling voice says: "I'm scared... it really bit me!" Administrator still protects her, warily watching pool. Background tourists whisper warnings: "Stay away, stop filming!" Ends with female protagonist shaken, crowd retreating, water ripples subsiding. [Live Audio] All live audio from phone, Chinese dialogue, no music, narration, subtitles. Start includes low tourist chatter, bamboo leaf rustling, slight water sounds. Administrator voice near and clear, shifting from serious warning to urgent restraint as danger occurs. During crocodile emergence: heavy water sound, female scream, nearby gasps; followed by phone splashdown, water crash, rapid panting, chaotic footsteps. Max splash and scream cause mic clipping briefly, then auto-gain recovery. Distant tourist voices more scattered, lighter, no monster roars added. [Continuity and Key Constraints] Opening phone must correctly film underwater target: screen up facing protagonist, rear lens down aimed at crocodile; external photographer sees mainly screen/sides. Prohibit rear lens facing photographer while depicting filming underwater target. First frame only blurry huge underwater shadow, no clear crocodile head/eyes/scales/teeth. Red liquid only after bite. Maintain continuity of Character #1, Outfit #2, Administrator, Crocodile, Boardwalk. Injury always same forearm, red streaks/fabric damage/tissue press location correspond. Only one phone, does not return to hand after fall. Protagonist/Administrator always inside railing, Crocodile @d57ec59a-c0a5-4eae-84f4-4f02ef21a92d always in pool. Maintain causal relationships of contact, weight, bite, release, retreat actions throughout. No wound close-ups, dismemberment, deep tissue display. Single continuous shot, no slow motion, replays, studio reveals, UI overlays, watermarks.

Generate video
Seedance 2.5
View full prompt & copy

CAMERA / LOOK: Handheld mini DV camcorder footage filmed by the subject herself. Slight hand shake, occasional focus hunting, imperfect framing, natural zoom adjustments, soft tape-like image quality, subtle grain, realistic auto-exposure shifts from soft bright garden daylight. Natural skin tones, mild motion blur, authentic consumer camcorder aesthetic rather than polished cinematic footage. STYLE: Cozy nature-care vlog with gentle ASMR elements. Relaxed pacing, minimal dialogue, candid moments, natural pauses. Focus on satisfying everyday sounds: mesh enclosure unzipping, soft rustling of leaves, gentle breathing, light footsteps on grass, faint birdsong. SUBJECT: Young woman in her mid-20s, plain linen dress, hair loose, minimal jewelry, no visible logos or branded items. Calm, quietly emotional energy during a small meaningful moment. SETTING: Small backyard garden on a bright calm morning. Natural daylight, flowering plants, a plain mesh butterfly enclosure resting on a table, no visible signage or brand names in frame. STORYBOARD: → (3s, propped medium shot) Places camera on a low table, gently unzips the mesh enclosure. "Finally ready to let this one go." → (3s, close-up) Reaches in carefully, lets the butterfly climb onto her finger. → (3s, handheld shot) Holds her hand up toward a nearby flower. "Take your time, no rush." → (3s, detail shot) The butterfly rests briefly on a petal before opening its wings. No dialogue. → (3s, medium shot) Watches quietly as it flutters a short distance away. "There you go." → (3s, macro shot) Wings catching the morning light as it lifts off. → (3s, propped shot) Follows it with her eyes until it disappears over the fence. → (3s, warm ending shot) Wipes a small tear, genuinely moved smile. "Didn't think I'd get emotional over this." → (6s, final shot) Turns toward camera. "See you next release." Hand covers lens as recording ends. AUDIO NOTES: Natural garden ambience — mesh unzipping, soft rustling, faint birdsong should be clearly audible. Dialogue quiet and casual, as if speaking to a small personal audience. REALISM NOTES: Authentic body language, natural blinking, genuine emotional smiles, occasional quiet pauses, imperfect framing, focus breathing, soft bright morning light shifts. No copyrighted characters, logos, brand names, or recognizable public figures anywhere in frame. Fully original personal vlog content, not AI-generated or commercial in style.

Generate video
Grok Imagine
View full prompt & copy

1990s camcorder found footage, VHS tape artifacts, tracking lines, chromatic aberration, auto-focus hunting, overexposed highlights, shaky amateur handheld camera, a date timestamp burned into the corner. Color grade: washed out gray-green, faded skin tones, blown out pale sky. A colossal dark lake creature with a ridged back and a long neck, mostly seen in silhouette through fog. A family on a wooden dock with a canoe at a northern lake. SHOT 1 (0 to 3s): Wide handheld shot matching the reference image, the camera zooms in clumsily as the creature's back breaks the surface far out on the lake, water rolling off in sheets. SHOT 2 (3 to 6s): Handheld medium shot, the camera whips toward the family on the dock, a child pointing, a father pulling the canoe out of the water, the image glitching with tracking lines. SHOT 3 (6 to 9s): Wide shot, auto-focus hunting, the long neck rises out of the fog, impossibly tall, a wave rolling toward the dock. SHOT 4 (9 to 12s): Running handheld shot, the camera bouncing as the operator flees up the wooden stairs, the lake visible in flashes behind, the wave slamming into the dock. SHOT 5 (12 to 15s): Static shot, the camera dropped on the grass and tilted sideways, the creature's silhouette slowly sinking back into the fog, then the tape glitches to blue screen. Loons calling, tape hiss, wind on the microphone, a deep foghorn-like call from the lake, a child shouting, panicked breathing, wood cracking, a final VHS tracking screech into silence.

Generate video
Seedance 2.5
View full prompt & copy

A single continuous 30-second take with no cuts. Rider, an adult male downhill longboarder, bombs a two-lane mountain road down a green Swiss alpine valley at extreme speed. The road is smooth grey asphalt with a white dashed centre line, a metal guardrail on the valley side, and white-and-black roadside marker posts. On one side, a steep bright-green meadow rises with old wooden Swiss chalets with red-brown shutters and dark timber barns. On the other side, dense dark conifer forest falls away into the valley. Forested mountains with green pasture clearings fill the distance, with low white clouds drifting across the peaks under a bright overcast sky. There are wooden utility poles along the road. [Rider] Rider wears a white full-face downhill helmet with a dark visor, so his face is never visible. He wears a plain white T-shirt, slim grey pants and black-and-white low-top skate shoes. He wears black slide gloves with round white hard pucks on the palms. He rides a long downhill board with black grip tape and bright red urethane wheels. [Image Quality] Ultra-high-definition 8K, with crisp, razor-sharp detail on Rider, his clothing folds, the glove pucks, the board and the red wheels. The asphalt grain, the meadow grass, the chalet wood and the distant tree lines are finely resolved. It has a clean, high-end action-camera look with no compression artifacts, no noise and no smearing. Rider stays sharp while the roadside streaks with strong directional motion blur. [Camera and Look] A low chase camera follows him like real raw-run footage: a wide-angle action lens about 1.5 meters off the ground, 3 to 4 meters behind, with natural high-frequency vibration from the road. The camera never stays dead-centre. It drifts left and right with each corner and keeps Rider on the left or right third, so the next corner and the open road ahead are always readable. The look is ultra-realistic documentary sports footage in natural soft overcast daylight, with no slow motion. [Speed Rule] Speed is everything in this video. From the first frame Rider is already travelling at over 100 km/h, and he never visibly slows down. Every corner is taken at full speed. His hand touches the asphalt only as a light, fast dab for balance and grip, never as a braking slide, and the board stays nearly straight. Immediately after each corner he snaps back into a deep tuck and keeps accelerating, so each straight is faster than the last. The motion blur, the white dashes flicking past rapidly and the wind tearing at his T-shirt never ease off. [Stage 1 / approx. 0–3s] Instant corner The video opens mid-run at full speed with a fast right-hand curve already arriving. Rider leans hard into it in a low crouch, dabs his gloved right hand on the asphalt with a quick scrape and sweeps through the apex without losing speed. The camera swings around the outside of the curve, and the guardrail whips past in the foreground. End state: Rider exiting the curve, still at full speed. [Stage 2 / approx. 3–10s] Acceleration straight Continue from the previous stage. Rider snaps into a deep aero tuck: front knee bent, back knee tucked behind the front calf, chest low, hands clasped behind his back. The road drops steeply down the valley. The chalets on the meadow side and the utility poles flash past one after another, and the white centre dashes strobe under the camera. His T-shirt ripples violently in the wind. End state: Rider at an even higher speed, with a left-hand curve rushing toward him. [Stage 3 / approx. 10–17s] Corner with oncoming car Continue from the previous stage. Rider dives into the left-hand curve at full speed, dabs his left hand on the asphalt and holds a tight inside line. Right at the exit, a white compact car appears coming uphill in the opposite lane and passes within a few meters in a heavy rush of wind. Rider stays low and committed in his own lane, without flinching or braking. The camera lets the car pass through the edge of the frame and then settles back behind him. End state: The car is gone and Rider is back in his tuck, still accelerating. [Stage 4 / approx. 17–23s] Right-left switchback Continue from the previous stage. A quick right curve flows straight into a left curve. Rider carves through both at full speed, with one fast hand dab on each side and his shoulders rolling with the road. The camera follows close, its horizon tilting slightly with each lean. End state: Rider exits onto the longest, steepest straight of the run. [Stage 5 / approx. 23–30s] Top speed Continue from the previous stage. Rider is in his deepest, tightest tuck, and this is the fastest moment of the whole video. The roadside smears into pure streaks of green and grey while the valley and mountains ahead stay steady. The wheels hum at a high pitch, and the board is perfectly stable. End state: Rider at absolute top speed, small against the huge green valley, still pulling away down the road. [Maintain Consistency] One rider only, always the same Rider with the same white helmet with the visor down, white T-shirt, grey pants, black gloves and red-wheeled board. His face is never shown. He always rides downhill in the same direction of travel. The car stays in the opposite lane and passes once. The take is one continuous shot with no cuts, no slow motion, no slowing down and no on-screen text. Audio is raw on-board sound only, with no music: <roaring wind buffeting the camera>, <constant hard rumble of urethane wheels on asphalt>, <quick gritty scrapes of the glove pucks on the road>, <a car engine growing louder then a sharp whoosh as it passes>, <the wheel hum rising in pitch as the speed climbs>.

Generate video
Seedance 2.5
View full prompt & copy

Create a 15-second ultra-realistic 16:9 landscape video featuring the same naturally attractive 24-year-old Korean woman, with long loose dark hair, minimal makeup, an oversized maroon hoodie, and loose casual pants. She is walking along a quiet European-style street in the early morning, carrying a long baguette in a brown paper bag. Preserve her exact face, outfit, hair, and body proportions across every shot. Shot 1 — 0–3 seconds: A medium-wide handheld shot from behind and slightly to the side. She walks casually along the sidewalk, holding the paper bag with the baguette sticking out. The camera follows her like a friend casually filming home video. Shot 2 — 3–6 seconds: The bottom of the paper bag suddenly tears. The baguette slips out, drops onto the sidewalk, and begins rolling away. She notices it and stops mid-step, looking down in surprise. Shot 3 — 6–9 seconds: A low-angle handheld shot follows the baguette rolling along the pavement. It wobbles realistically, briefly bumps against a small uneven patch, and continues rolling. Her shoes enter the frame as she hurries after it. Shot 4 — 9–12 seconds: A quick, imperfect camera reframe catches her bending down and reaching for the baguette just as it rolls a little farther away. She makes a small, playful lunge and finally grabs it with both hands. Shot 5 — 12–15 seconds: A slightly crooked medium shot captures her standing up, inspecting the baguette, then looking toward the camera with an embarrassed, amused smile. She hugs it against her hoodie as if she has rescued something precious, then laughs naturally. Visual style: Authentic early-2000s Sony MiniDV home-video footage, filmed by another person, not a selfie. Slight handheld shake, imperfect framing, natural camera corrections, occasional autofocus hunting, subtle exposure shifts, faded colors, soft contrast, realistic skin texture, and ordinary morning street lighting. The moment should feel spontaneous, cute, and unintentionally funny rather than staged or cinematic. Motion and physics: Natural human movement, believable gravity, realistic rolling and friction, continuous real-time motion. No floating baguette, no impossible bouncing, no sudden teleportation, no duplicated objects, no stuttering, no judder, no frame skipping, no slow motion, and no exaggerated cartoon reactions. Audio: Natural street ambience, footsteps, paper bag rustling, the baguette softly hitting the pavement, and her genuine little laugh. No music, narration, or artificial sound effects.

Generate video
Seedance 2.5
View full prompt & copy

Invited by hand gesture, puts on helmet. Pushed back by the wind when the door opens, caught on the back. Jumps in and body floats up, but is swept sideways and shoulder touches glass. Waist grabbed and returned to center, sinks almost to the wire mesh. Arms spread wide, gently lifted. Supported by shoulders, floating horizontally. Signal to extend legs, signal to lift chin. Even after hands are released, spreads arms again to float alone. Thumbs up given, mouth relaxes into a smile. Waist grabbed, two people spiral up to the top of the tube. Laughs with mouth open at the top. Comes down and passes through the entrance/exit, knees buckle and stumbles. Takes off helmet, hair is messy, laughs and high-fives. Recorded handheld by friend from outside the glass. Approx 2 seconds per cut, no dialogue/subtitles/BGM. Always prone while floating, feet do not touch the wire mesh. Only black-framed doors for entry/exit. Uniforms and helmets are plain.

Generate video
Seedance 2.5
View full prompt & copy

Autumn grassland on top of a mountain. Wings rise overhead, pulled backward along with the harness, shifting half a step. Just after regaining footing, the helmet slips down over the eyes, pushed up with one hand. Start walking late, shoved from behind by the back, stumbling forward. Small steps gradually widen, feet kicking grass become light, soles leave the ground. Already floating, yet legs are still running two or three steps. Gripping the riser tightly, failing to sit properly on the seat, settling firmly on the second attempt. Hair tips touch cheeks, shaking head. Both hands gently release, handed an orange handle. Pulling strongly to the right tilts the horizon, stopping hands. With help holding hands, pulling slowly again, river below crosses horizontally. On landing, this time running feet properly catch the ground. Deflated wings cover back, laughing while shaking hands. Handheld shot from friend's perspective. Approx 2 seconds per cut, no dialogue/subtitles/BGM. One wing, two harnesses, two helmets remain constant throughout. Sakura is always in front, instructor always behind.

Generate video
Loading more prompts

FAQ

Where do YouMind prompts come from?
YouMind curates high-signal AI prompt examples from creator communities and public creative workflows. The library focuses on prompts that are useful to inspect, reuse, and adapt.
How are video prompt categories organized?
Video prompts are grouped into use cases, styles, and subjects. Use cases cover the production goal, styles cover motion or rendering language, and subjects describe the main scene or entity.
How are prompts assigned to categories?
YouMind uses AI to analyze each video prompt, its title, metadata, and generated result when available, then assigns the closest categories. Editors can refine categories as the taxonomy evolves.
What is YouMind?
YouMind is an AI workspace for turning research, ideas, and creative references into usable outputs. The prompt library is one public part of that workflow, built to help creators learn from strong examples faster.