jtydhr88/ComfyUI-InstantMesh

Two steps process

Abocg opened this issue · 1 comments

Hi. Thanks for your implementation. I just wanted to ask if possible to make it two steps process so that I can refine the generated multiview image in stable diffusion and give it more details and feed it back to the 3D processer just like CRM code in Mr.forexample 3D pack. Many thanks in advance

hi @Abocg, in fact InstantMesh has plan to support different Diffusion model but not finish yet, see
image
https://github.com/TencentARC/InstantMesh
So once they complete, I will implement/update this comfyUI custom nodes here.
Please keey an eye on it. I will close this issue for now.