search

aim-uofa / GenPercept

GenPercept: Diffusion Models Trained with Large Data Are Transferable Visual Models
https://huggingface.co/spaces/guangkaixu/GenPercept
Creative Commons Zero v1.0 Universal
91 stars 2 forks source link

How to encode pose? #1

Closed chester256 closed 2 months ago

chester256 commented 2 months ago

Thank you for the code. How do you encode the 17-channel heatmaps to 3 channels to fit the VAE?

YongtaoGe commented 2 months ago

Hi, we encode the heatmaps with predefined colormaps, which is similar to Painter. You can have a reference from here.