multi-subject-render VS depthmap2mask

You can use the multi subject renderer https://github.com/Extraltodeus/multi-subject-render.git

Get this Extension, and as always, please read the docs to avoid problems.

There's also a script/extension https://github.com/Extraltodeus/multi-subject-render but it's fiddily to get work right, and i think the other workflow is faster.

There are some ideas to create multiple different subjects, such as this extension for automatic (https://github.com/Extraltodeus/multi-subject-render), or Area Composition if you are using ComfyUI (https://comfyanonymous.github.io/ComfyUI_examples/area_composition/).

you can do it with https://github.com/Extraltodeus/multi-subject-render but i don't really know how to use it

There is an extension called multi-subject-render that allows you to provide one prompt for the background and a second prompt for the foreground.

for change background i suggest to use depth2mask.

have you try with dept2mask?

Looks really good. Have you tried img2depth for the texturing? GitHub - Extraltodeus/depthmap2mask: Create masks out of depthmaps in img2img

have you try https://github.com/Extraltodeus/depthmap2mask ?

No, and I don't need to. depthmap2mask works sloppy, I don't like it. It's much better to create mask for "Inpainting" using image-editing software. Here you can see how it's done. https://www.youtube.com/watch?v=dnIYTGW1m8w

Sent to img2img and used Depth Aware img2img mask with the model set to `midas_v21_small` so that I would hopefully affect as little of the image as possible. (after seeing the pants morph, I think it might have been better to just use inpaint)