search

ziqihuangg / Collaborative-Diffusion

[CVPR 2023] Collaborative Diffusion
https://ziqihuangg.github.io/projects/collaborative-diffusion.html
Other
405 stars 31 forks source link

about segmentation masks #39

Open jupytera opened 4 months ago

jupytera commented 4 months ago

Hello, I think your work is excellent and inspiring. May I ask if you have the code to convert facial images into segmentation masks? Thank you very much!

ziqihuangg commented 4 months ago

Hi, we used the segmentation masks provided in the CelebAMask-HQ dataset: https://github.com/switchablenorms/CelebAMask-HQ They also provided a face parsing network that you can use.

jupytera commented 4 months ago

Thank you very much for your answer. Therer is another question I am eager to get answer. During training, I generated mask_gs.png as shown. Is this intended to help the model learn completion by erasing images at fixed positions?Thank you for your help again. `$1HA )UU~XOFKQY` 6YOX4