1 / 4

esc

1 / 4

FilmLune

Seedance

PROMPT

Create a vertical 9:16 photorealistic opening image for a miniature fashion video.

A clearly adult East Asian woman, approximately 28 years old, with long wavy dark-brown hair, is in a warm cream-colored living room with paneled walls, wall sconces, sheer white curtains and a light-colored rug.

Show her full body from head to bare feet. She is miniature-sized, with one giant realistic hand entering from the right and gently supporting her upper back. Her feet hover just above the rug, ready to be lowered onto it.

She wears a loose white off-the-shoulder button-front pajama blouse and short white lace-trimmed pajama bottoms. The blouse is still on.

Soft natural daylight, realistic fabric, a calm expression, and a fixed front-facing camera. No floral dress, shoes, necklace, hair clip, handbag, text, or logo.

Create a vertical 9:16 photorealistic intermediate-outfit reference image for a miniature fashion video.

The same clearly adult East Asian woman, approximately 28 years old, with long wavy dark-brown hair, stands upright in the same warm cream-colored living room with paneled walls, wall sconces, sheer white curtains and a light-colored rug. Keep her face, body proportions, lighting and front-facing camera angle consistent with the opening image.

The miniature adult woman wears an opaque white strapless bra and the same short white lace-trimmed pajama bottoms. The white outer blouse has been taken off her shoulders and hangs loosely around her forearms, held by two giant hands at the sides. No nudity.

Show her full body from head to bare feet, with a natural, neutral pose. Soft natural daylight, realistic hands and fabric. No floral dress, shoes, accessories, text, or logo.

Create a vertical 9:16 photorealistic final-outfit reference image for a miniature fashion video.

The same clearly adult East Asian woman, approximately 28 years old, with long wavy dark-brown hair, stands facing the camera in the same warm cream-colored living room with paneled walls, wall sconces, sheer white curtains and a light-colored rug. Keep her face, body proportions, lighting and camera angle consistent with the opening image.

She is fully dressed in a pink-and-ivory floral dress with bow shoulder straps, a fitted waist, and a softly layered skirt. She wears a small flower pendant necklace, a flower hair clip, white mid-calf socks, and cream Mary Jane shoes. She holds a small cream handbag with a bow.

Show her full body from head to shoes. Natural proportions, realistic fabric, soft daylight, and a gentle expression. No giant hands, extra garments, text, or logo.

Generate the three reference images with their own prompts, then upload them to the video step in this order: image 1 (opening outfit), image 2 (blouse removed), image 3 (finished outfit). The video prompt calls them @Image 1, @Image 2 and @Image 3.

Create a single 15-second vertical 9:16 photorealistic miniature fashion-doll dress-up video featuring the same adult woman throughout. This is a neutral fashion demonstration with no nudity.

Reference roles: @Image 1 is the opening-state reference: begin with the same woman, white off-the-shoulder pajama blouse and white shorts, giant hand, room, lighting and pose shown there. @Image 2 is only the intermediate clothing state after removing the white outer blouse; preserve the same woman from @Image 1. @Image 3 is only the finished pink floral dress and its accessories; do not introduce a second woman or room. These are three reference images, not pinned keyframes or a last frame.

There is ONE outfit change, from the white pajamas to the floral dress. There is exactly ONE floral dress in the scene. It is one continuous physical garment: two giant hands lift it, put that same dress on the woman, then release and smooth it. Never show a duplicate dress in another hand or next to her. At most two giant hands are visible at once. No instantaneous clothing swaps, extra people, or new garments appearing from nowhere.

0–2 seconds: The giant hand lowers the miniature woman from the pose in @Image 1 until both bare feet touch the rug. Gently turn her slightly and return her to the front.
2–4.5 seconds: The giant hands gently open and remove only the white outer blouse. The same white shorts remain in place; her opaque white bra and shorts match the intermediate state in @Image 2. Move the removed blouse out of view.
4.5–7.5 seconds: Two hands bring ONE pink floral dress from @Image 3 into the frame, place that same dress on her over the base garments, set the shoulder straps and smooth the skirt. No second dress appears.
7.5–9.5 seconds: Brief foot close-up. The hands put on the white socks, then the cream Mary Jane shoes, one foot at a time, before returning to the full-body view.
9.5–12.5 seconds: The hands place a small flower necklace, a flower hair clip, and then a cream handbag, one after another. Show each placement and release.
12.5–15 seconds: The giant hands withdraw. The woman turns once in the completed outfit and faces forward as her hair and skirt settle.

Preserve the same adult woman's identity and the same cream-colored room throughout. Realistic anatomy, natural cloth physics and warm daylight; fixed full-body camera except for the brief foot close-up. Light instrumental music and subtle fabric sounds, no dialogue, captions or logos.

Prompt guide · AI video

Giant hands dress a miniature woman — AI Video Prompt

Two giant hands take a white pajama blouse off a miniature adult woman, dress her in a pink floral dress and add socks, shoes and accessories in one 15-second clip.

ModelSeedance 2.5
Use caseFashion & beauty
StyleWarm miniature dress-up
CollectionOutfit change videos

One continuous outfit change from three ordered reference images

A giant hand lowers a miniature adult woman in white pajamas onto a rug. Two hands open and remove the white blouse, leaving an opaque white bra and shorts; there is no nudity. They bring in one pink floral dress, put it on her and smooth the skirt. After a close-up of white socks and cream Mary Jane shoes, the hands add a flower necklace, a hair clip and a cream handbag. The hands withdraw and she turns once in the finished outfit. Each reference image fixes one state of the same woman: the opening outfit, the base garments and the finished look. The prompt names each image's role and says they are references, not keyframes, so the model moves between them instead of jumping. Timed beats, a single dress and at most two hands reduce duplicated garments and sudden swaps.

How to adapt the method

Change the outfits in the three image prompts while keeping the same woman, room and lighting in all three images. Keep the intermediate image dressed in opaque base garments. Upload the images in the same order and update the clothing in the video prompt so each @Image reference still matches its role. For more or fewer steps, change the reference list and the timed beats together. Adapted image prompts with separate source images removed; the video prompt is the one used for this clip.

How to use this prompt

1
Generate the three reference images with their own prompts, then upload them to the video step in this order: image 1 (opening outfit), image 2 (blouse removed), image 3 (finished outfit). The video prompt calls them @Image 1, @Image 2 and @Image 3.
2
Open the selected recipe in Studio with its model and starting settings. Check the available version, frame and resolution before generating; a size written in the source prompt is not an output setting or a guarantee.
3
Use the customization fields below to rewrite the brief: Reference images / Delivered dimensions / Delivered duration. Keep the composition and other constraints that make this example useful. Check the current credit quote before submitting.

What to check in your result

Watch the entire clip. Compare the main action, camera movement and scene continuity with the brief, and check for changing faces, distorted hands or unexpected cuts. A similar prompt can give a different result; refine the motion rather than assuming an exact copy.

Variables to customize

Reference images: Three ordered images: opening outfit, blouse removed, finished outfit
Delivered dimensions: 720 × 1280
Delivered duration: 15.041667s

Source and attribution

This example is attributed to FilmLune.

View the source on the public source

Report a rights or attribution issue

View all AI video examples

Main Character City Walk

Main Character City Walk