Happy Dog and His Family | Live-Action Video Breakdown (Prompts Included)

Upload your prepared image or pick assets from your Creations library. Select reference-to-video mode and paste your video prompt. You can also upload a video for motion reference.

Note: Add @your-image for image reference and @your-video for video reference inside the prompt. Pick any of these models: Seedance 2.5, Seedance 2.0, MiniMax H3 or Wan 3.0. Click Generate after confirming all settings.

Video Prompt: Five-Scene Winter Family Montage

PROMPT HEADER:

Generate a 30-second horizontal 16:9 1080p live-action winter montage with five distinct scenes. Use @Reference image as the global cast reference and @Reference video as the motion, pacing, and camera-rhythm reference. Create a photorealistic live-action family winter film with natural child performances, nostalgic warm color, subtle 35mm grain, realistic snow, realistic clothing, and stable continuity.

GLOBAL CAST REFERENCE:

Use @Reference image for character appearance and clothing. The cast is fixed:

CHARACTER A = the blond boy wearing the white cable-knit sweater, blue collar, orange neck scarf and blue jeans.

CHARACTER B = the petite girl wearing the orange turtleneck and red skirt, with short dark-auburn hair, square black glasses and visible freckles across her nose and cheeks.

CHARACTER C = the tallest lanky boy wearing the faded green layered shirt and brown trousers.

CHARACTER D = the red-haired girl wearing the purple dress, purple headband, green neck scarf and lavender tights.

DOG = the large brown Great Dane wearing the blue collar.

Keep A, B, C, D and DOG visually identical to @Reference image throughout the full video. Never exchange their clothing, hair, faces, positions, or identities. Show exactly four children and one dog.

REFERENCE VIDEO USAGE:

Use @Reference video only for natural winter montage pacing, camera movement, child performance rhythm, casual group energy, and realistic motion continuity. Do not copy unrelated background elements, extra people, text, captions, logos, or character changes from @Reference video. The scene plan below controls the final video.

SCENE 1 — 00.0 to 06.0 — BURGER KIOSK:

Location: snowy roadside burger kiosk at dusk.

Camera: begin with a wide establishing shot, then cut to a medium side angle.

A stands first in line and receives a plain wrapped burger from the service window.

B stands directly behind A, counts coins with gloved hands, then pushes her black glasses up; her freckles remain visible.

C stands third, studies a simple picture menu with no readable words, then points excitedly at a burger picture.

D stands fourth and holds the DOG's blue leash with her left hand.

DOG sits calmly beside D and looks toward the food.

Do not add any other visible customers or workers. Only a pair of hands may appear briefly through the service window.

SCENE 2 — 06.0 to 12.0 — CLEARING SNOW:

Location: entrance of a small community recreation building.

Camera: wide side-tracking shot moving slowly from left to right.

A uses a full-size snow shovel to push one narrow strip of snow away from the steps.

B uses a small child-sized shovel to clear snow near the doorway; glasses and freckles remain unchanged.

C brushes snow from the van windshield with a long soft brush.

D holds the DOG's leash and sweeps light snow from a bench with a hand broom.

DOG watches B's shovel and playfully catches a falling snowflake.

All actions are slow, safe, realistic, and non-overlapping.

SCENE 3 — 12.0 to 18.0 — EATING INSIDE THE VAN:

Location: warm van interior.

Camera: stable medium-wide shot followed by two short close-ups.

A sits in the front passenger seat holding a plain wrapped burger.

B sits behind A eating fries from a plain paper tray; she laughs softly and adjusts her glasses.

C sits opposite B holding a burger with both hands.

D sits beside C and passes one fry to B.

DOG sits on the floor between B and C, watching the food without jumping.

No branded wrappers and no readable packaging.

SCENE 4 — 18.0 to 24.0 — BUILDING A SNOWMAN:

Location: open snowy park in daylight.

Camera: wide frontal shot with one brief close insert.

C carefully places the snowman's round head onto the body.

A inserts the left and right stick arms.

D wraps a plain blue scarf around the snowman.

B presses a carrot nose into the center of the face; her orange sweater, glasses, and freckles remain clearly visible.

DOG circles the finished snowman once, then sits beside B.

Each character performs only the assigned action.

SCENE 5 — 24.0 to 30.0 — PACKING AND GROUP PHOTO:

Location: beside the van at blue hour.

Camera: medium-wide rear three-quarter shot, then slow pullback.

A places two snow shovels into the rear cargo area.

C lifts a small lightweight sled and slides it beside the shovels.

B closes a plain paper food bag and places it inside the van.

D positions a small instant camera on a low tripod, activates the timer, then joins A, B and C.

A stands on the left, B stands second, C stands third, D stands on the right, DOG sits in front.

The camera flash fires once; all four children laugh naturally; end with a slow pullback.

VISUAL RULES:

Photorealistic live-action family winter film, natural child performances, nostalgic warm color, subtle 35mm grain, realistic snow and clothing. A, B, C, D and DOG remain stable across every cut. No adults, no extra children, no duplicate characters, no missing cast member, no identity swaps, no face changes, no clothing changes, no morphing, no dangerous actions, no readable text, no captions, no subtitles, no logos, and no watermark.

AUDIO:

Continuous upbeat original winter-comedy music. Add burger-wrapper rustle, quiet kiosk ambience, shovel scraping, snow brushing, van-door sounds, children laughing naturally, dog panting, and one gentle bark. Include brief natural English dialogue without subtitles.

NEGATIVE PROMPT:

Extra customers, visible workers, adults, extra children, duplicate children, duplicate dog, missing child, missing dog, identity swap, clothing swap, changed hair, changed faces, changed glasses, missing freckles, missing blue dog collar, wrong dog breed, character morphing, face drift, wardrobe drift, screen-position identity exchange, unsafe shoveling, dangerous action, overlapping actions, unreadable character faces, branded food wrappers, readable menu words, readable packaging, captions, subtitles, logos, watermark, random text, title cards, extra vehicles, harsh horror lighting, dark thriller tone, slapstick injury, aggressive dog behavior, jumping dog, snowman action assigned to wrong character, wrong final lineup order, camera flash repeating, low resolution, blurry faces, excessive motion blur, distorted hands, extra limbs, broken leash, unrealistic snow, frozen characters, unnatural laughter, audio mismatch, missing winter-comedy music, missing natural sound effects.

Final Summary

Create one complete 30-second horizontal 16:9 photorealistic winter family montage using Reference image as the fixed appearance and wardrobe source for exactly four children and one Great Dane. Character A, Character B, Character C, Character D, and DOG must retain the same faces, hair, clothing, scale relationship, and identities through the burger kiosk, snow clearing, van meal, snowman building, packing, and final group photo scenes.

Use Reference video only for natural motion, camera rhythm, and performance pacing. The written five-scene prompt controls the exact locations, actions, cast positions, final lineup, visual restrictions, and audio requirements. The finished video should feel like a nostalgic live-action winter family short with warm comedy music, realistic snow, plain unbranded props, no readable text, no extra people, no identity swaps, and no watermark.