Toddler in Blizzard with Distant Wolf Howls, Hand-Drawn 2D

A hand-drawn 2D animation on twos depicting a toddler's face in a violent blizzard, with wolves appearing only as a distant dark band on the horizon. The prompt emphasizes frame-by-frame micro-acting, handheld camera work, and a flat, directionless ambient light.

This clip could not be loaded — it may have been removed from X.
オズ @ozuozuai99 ·
View on X ↗

Prompt

Style: 8K IMAX, traditional hand-drawn 2D animation, animated on twos at 12 frames per second — each drawing held for two frames then replaced, choppy stepped motion cadence, hand-painted oil-brush texture, brushstrokes shifting and redrawn from frame to frame, line jitter, no 3D render, no game engine, no CGI smoothness.

DIRECTOR'S NOTES — non-negotiable for every frame and every cut:
1. CHARACTER REFERENCE IS ABSOLUTE — face from <<<image_3>>> and wolf design from <<<image_4>>> exactly 1-to-1, no deviation, reference always overrides text.
2. MICRO-ACTING ON TODDLER FACE — every frame of her close-up shows constant micro-expressions: involuntary blinks as distinct held closed-eye frames on twos, sustained hard squints under gusts, eyelid flutter between blinks, brow gathering, chin quivering, lip pressed and trembling, jaw small and tight, nostrils flaring with shallow breath, tiny scared saccades searching the white air, eyes welling but no tears fall. When the first wolf howl rises offscreen, her eyes snap open wide for a single held frame in pure shock, eyelids freeze, breath catches in her small throat, lip parts slightly.
3. WOLVES NEVER APPROACH — they appear ONLY as a distant dark cloud creeping along the far horizon, never close, never as individual animals, only an undulating dark band of forms — appearing and vanishing as the buran sweeps across the lens.
4. BURAN NEVER PAUSES — rolling white-out waves continuously in every frame.

CHARACTER TAGS:
- THE TODDLER = small girl from <<<image_3>>>.
- THE WOLVES = animals from <<<image_4>>>.
- THE FOREST = location from <<<image_1>>>.

Cinematography Lubezki / Deakins, fully handheld in human hands breathing with operator, anamorphic wide lens, shallow depth, never tripod never dolly never crane never aerial, eye level or below.

The 12 principles of animation throughout: anticipation, follow-through and overlapping, slow in slow out on head turns, arcs, secondary action, exaggeration, solid drawing.

Lighting: night, no light source, no moon, no stars, no rim, no contre-jour, no key light. Flat dim diffuse ambient grey glow only, no direction. Flat painted shapes, no gradient shading.

Atmosphere — violent buran continuous: hurricane gusts driving snow horizontally, visibility roughly 3 meters around the camera, rolling waves of denser blizzard sweeping the lens, snow piled on the toddler's cap and shoulders, wind never drops.

Audio: roaring buran wind continuously as dominant bed, crunching snow, shallow ragged child's breathing, plus two long wolf howls rising from offscreen distance — one howl in the first half of the scene, a pause, then a second howl from a different point along the horizon answering the first in the second half. Distant low growls and panting from the dark mass barely audible under the wind. No dialogue. No music. No subtitles.

SHOT 2A — TIGHT CLOSE-UP handheld on THE TODDLER's face, exact face from <<<image_3>>>, every feature 1-to-1, identical features identical hair identical cap. Camera very close at her eye level, breathing with operator, vertical sway, her small face filling the frame off-center, negative space of white-out, focus locked on her eyes, background dissolved into pure white-out blur. Flat painted shadow only, no light direction. Snowflakes catch on her dark lashes and on the fur of her cap, snow piling on her shoulder.

MICRO-ACTING in this shot, drawn frame by frame on twos as held poses: she squints hard against the wind, eyelids fluttering, then a held closed-eye frame for two frames, then snapping open, then a sustained hard squint when a gust hits held for three frames, anticipation as her tiny brow gathers before each squint, slow in slow out on every eye closure. Chin quivers visibly between beats, lower lip pressed and trembling, jaw small and tight, nostrils flaring with each shallow ragged breath, eyes welling at the corners but no tears fall.

Then the first wolf howl rises from offscreen — her eyes snap wide open for a single held frame in pure shock, eyelids freeze in that wide pose for two frames, breath catches in her small throat (visible by held still chest), lip parts slightly. Her gaze then shifts in tiny scared saccades searching the white air on arcs, head turns a fraction toward the sound on a slow arc. The second howl answers from a different direction — she flinches, brow tightens, another sharp involuntary blink as a held closed-eye frame. A wave of buran sweeps across the lens partly obscuring her face, she does not move. Painting style identical to references. Animated on twos.

HARD CUT to SHOT 2B — VERY WIDE LOW handheld looking out across the snowy expanse beyond THE FOREST treeline, the operator crouched in snow at knee height looking out toward the far horizon, camera in their hands with visible breath sway, the nearest dark pine trunks framing the edges as flat silhouettes.

Far in the distance at the very edge of visibility along the horizon, THE WOLVES move as a single body — a dark moving mass like a black cloud creeping along the snowline, never approaching, never resolving into individual animals, only an undulating dark band of forms — exact wolf design from <<<image_4>>> readable only as silhouettes, exact coal-black fur with cold steel-blue sheen, faint pale yellow eye pinpricks barely visible.

A rolling wave of buran sweeps across the lens — the dark cloud of wolves vanishes completely into white-out, holds white for two beats, then ghosts back into view as a faint dark band shifting along the horizon. The second wolf howl rises from a different point along the horizon, the dark band shifts in response stretching laterally. Another wave of buran swallows them again. The camera stays low and still, operator only breathing. Painting style identical to references. Animated on twos.

Constraints: two distinct shots with hard cut between them not one continuous take, TIGHT CLOSE-UP on toddler then HARD CUT to VERY WIDE on wolves, THE TODDLER's face matches <<<image_3>>> exactly 1-to-1 with constant frame-by-frame micro-acting drawn as held poses on twos, THE WOLVES remain a distant dark cloud along the horizon throughout and never approach or resolve into individual close animals, reference always overrides text, buran rolling waves present in every frame, no dialogue, camera handheld eye level or below, no light source, flat painted shadow only, no 3D look.
? How to use this prompt ▾
  1. Copy the prompt using the green "Copy prompt" button above.
  2. Open your model — launch Seedance 2.0 on its official interface.
  3. Paste & generate — keep the verbatim prompt text for best adherence. Set aspect ratio to 9:16 and clip duration to 15s.

What makes this prompt work

Every insight below is verified against benchmark generation parameters, not asserted. How our prompt engineering benchmarks work →

CAMERA
"dolly" is a core cinematic camera technique for establishing spatial immersion and subject framing. Also present: "crane", "handheld", "aerial/drone".
LIGHTING
"silhouette, soft light" establishes luminance depth, volumetric shadow falloff, and tonal contrast across the generation.
MOOD
"tense, dramatic" — Mood and aesthetic tone are explicitly directed to maintain emotional consistency throughout the scene.
DURATION
At 15s, this clip is optimized for high temporal coherence and fluid, realistic motion physics.
MODEL
Seedance 2.0 (ByteDance) provides the underlying diffusion and motion architecture. Camera, lighting and duration cues transfer across models — what differs is how literally each model interprets motion.
PROMPT LANGUAGE
The prompt is written in English. This library keeps every prompt in its original language rather than translating it — the exact words are what was conditioned during generation.
Seedance 2.5 BytePlus Cloud Partner

Tell bigger stories with Seedance 2.5

Create AI videos up to 30 seconds long with richer image, video, and audio references. Ideal for complex ideas that need consistency and control.

Explore on BytePlus Official ByteDance Enterprise Cloud