stable-diffusion-webui icon indicating copy to clipboard operation
stable-diffusion-webui copied to clipboard

Published 20 hours ago • verified publisher icon AUTOMATIC1111

[Feature Request]: Picture to 3D Model

Open pied-stuff opened this issue 2 years ago • 1 comments

What would your feature do ?

Convert a picture to High-Resolution 3D Human Model 3DHuman https://github.com/facebookresearch/pifuhd PIFuHD: Multi-Level Pixel-Aligned Implicit Function for High-Resolution 3D Human Digitization (CVPR 2020)

Proposed workflow

  1. Go to convert to 3D Human Tab
  2. Press convert image after import image
  3. Done

Additional information

No response

pied-stuff avatar Jan 27 '23 22:01 pied-stuff

Just use this, standalone. https://github.com/openai/point-e turn the notebooks to .py files and run it, also it works on windows with a small change but you can run this on WSL with 4gb vram It needs more vram for the img 2 3d model mode though. But this seems out of scope as built-in

ClashSAN avatar Jan 27 '23 22:01 ClashSAN

I still would love to see pifuhd as extension. I do believe it would something amazing to have.

oivio avatar May 18 '23 16:05 oivio