How to Insert and Maintain Specific Objects or Characters Across Generated Images in Diffusion Models

Models
1

Question:

I am working on a project using a diffusion model, and I would like to achieve the following:

  1. Is there a way to insert an existing image (for example, a character) into a newly generated image? I would like to specify aspects like the character’s pose or clothing via prompts.

  2. Once an image is generated, is there a method to keep specific objects (for example, a desk or chair) in the same form in future generated images? I want to significantly change other aspects of the image, but keep the object (the desk or chair) with possible variations in position or angle.

Any guidance or recommended approaches would be greatly appreciated!

3

thx! i’ll check it out.