cvlab-columbia / pix2gestalt

Code for the paper "pix2gestalt: Amodal Segmentation by Synthesizing Wholes" (CVPR 2024)
https://gestalt.cs.columbia.edu/
Other
134 stars 10 forks source link

Question about the RGB / RGBA output format #11

Open JingyeChen opened 4 months ago

JingyeChen commented 4 months ago

Thanks for your excellent work and it indeed inspired me a lot! I am wondering is it possible to generate completed images in RGBA format for easier segmentation?