DreamOmni2: 6 Before/After Image Edits

DreamOmni2 is an image editing model. It takes one or more reference images and a short instruction. The tests below focus on practical multi-image edits: outfit swap, scene composition, pattern transfer, and style transfer.

Model link

https://wiro.ai/models/xiabs/dreamomni2

Settings used

Inference steps: 25
Guidance scale: 3.5

1) Outfit swap (two images)

Inputs	Output
Input A Input B	Prompt: Replace the suit in the first image with the clothes in the second image.

This test checks identity and pose preservation. The hard part is keeping facial structure while changing clothing.

2) Two-character composition (two images)

Inputs	Output
Input A Input B	Prompt: In the scene, the character from the first image stands on the left, and the character from the second image stands on the right. They are shaking hands against the backdrop of a spaceship interior.

Multi-subject composition usually fails on scale and perspective. A good result keeps lighting consistent across both subjects.

3) Pattern transfer (two images)

Inputs	Output
Input A Input B	Prompt: Make the car in the first image have the same pattern as the mouse in the second image.

This test checks texture mapping. Look for pattern stretch or inconsistent reflections on curved surfaces.

4) Merge subjects + match style (three images)

Inputs	Output
Input A Input B Style reference	Prompt: The cat from Image 1 and the dog from Image 2 are sitting side by side, with the background inside a car. The style of the image is the same as in Image 3.