WeShop AIWeShop AI
scroll left

AI Image

Effects

AI Video

Pricing

20% OFF

Solutions

User Guide

Resource

Download

Affiliate

API

scroll right
Sign In
Resource
Explore guides and tips that help you get things done faster.
blog
Blog
blog
FAQ
feature request
Feature Request
jobs
Jobs
app image
Create Anytime, Anywhere with WeShop AI
Unleash your creativity on-the-go or at your desk. Download WeShop AI and turn inspiration into stunning designs, whenever and wherever it strikes. Powerful, flexible, and always at your fingertips.
November 12

Wan 3.0 Fashion Video Background Change: Take a Runway Walk Outdoors

Replace a runway background with Wan 3.0 using a short single-person walking shot. Compare the source and courtyard result for face, suit and action changes.

A useful background edit needs the performer to remain recognizable. This Wan 3.0 Reference test starts with a short medium runway shot of one man in a dark suit and white shirt. The request places him in a quiet outdoor stone courtyard while retaining his visible appearance, clothing and walking action.

Original input

Original medium runway shot of a blond man in a navy suit.

AI-generated result

AI-generated stone-courtyard setting with the recognizable source performer.

Keep one performer and one continuous camera move

Use a single-person section rather than a montage of several models. Here, the selected four seconds show a medium view with a readable face and suit. The original downloaded version is 540×960; it was resized to 720×1280 for input, which does not recover missing facial detail.

Adapt the new place to the original performer

The prompt asks for soft neutral daylight in an outdoor stone courtyard and keeps the existing light on the man. A simple courtyard gives the background one clear job. Asking for dramatic relighting, a new outfit and a different camera at the same time would make preservation harder to evaluate.

Tested prompt and reference roles

Edit Video 1 by replacing only the environment behind the single adult male runway performer. Put the same walking man in a quiet outdoor stone courtyard under soft neutral overcast daylight. Preserve his recognizable face, blond hairstyle, skin tone, expression, head shape, dark navy suit jacket and waistcoat, white open-collar shirt, trousers, hands, body proportions, posture, walking rhythm, shot size and camera tracking exactly as in the source. Keep the original light on the man; adapt the courtyard to him rather than relighting or regenerating his face. Remove the runway audience and dark indoor stage from the background only. Do not change the performer, suit, shirt, face, hair, body, action, timing or framing. No new foreground objects, duplicate people, new camera moves, added captions, logos, dialogue or music.

Judge the face and suit as strictly as the background

Compare facial appearance, blond hair, white open-collar shirt, dark waistcoat and jacket, then inspect the arm movement and shot size. Background replacement alone is not sufficient if the result becomes a different person. Treat any small detail reconstruction as a limitation rather than describing the result as unchanged original footage.

Decide where generative editing is acceptable

This workflow creates a new visual interpretation of a location. It is not a pixel-exact foreground composite. If a project requires the original face, clothing pixels and movement to remain untouched, a result with reconstructed details does not demonstrate that standard. Keep the same source while refining the environment instruction so comparisons remain meaningful.

Recorded settings for this example

Open Wan on WeShop, choose Wan 3.0 Reference, and add the original input. Paste the prompt, generate one result and inspect it against the source.

SettingTested value
WorkflowVideo-to-Video
Image inputs0
Video inputs1
Duration4s
Resolution720P
Aspect ratio9:16
Quantity1
Generate audioOff

What the selected result shows

  • Outdoor stone courtyard replaces indoor runway and spectators
  • Original blond male appearance, navy jacket/waistcoat, white open-collar shirt, hand-in-pocket stance and walking progression remain recognizable

Light and fine facial/clothing detail reconstructed; not pixel-exact foreground masking. Original video is 540x960 resized to 720x1280; resizing adds no true detail. These are selected outputs generated and reviewed on November 12, 2026; they do not measure repeatability across many runs.

You can try it directly using the tool on the right.

Reference credit: PB on Pinterest. Pinners are credited as source curators where original authorship is not established. This independent AI-editing demonstration does not imply endorsement by the source creators or depicted people. Use media and likenesses you are authorized to edit.