search

MoyGcc / vid2avatar

Vid2Avatar: 3D Avatar Reconstruction from Videos in the Wild via Self-supervised Scene Decomposition (CVPR2023)
https://moygcc.github.io/vid2avatar/
Other
1.23k stars 100 forks source link

Export extra data? #60

Open antithing opened 10 months ago

antithing commented 10 months ago

Hi, and thank you for making this project available. Will it work with moving cameras? Is it possible to export the clean background as rgb frames, and the segmentation masks as frames, as well as the animated 3d mesh?

Thanks!

MoyGcc commented 10 months ago

Hi, thanks for your interest. Yes, it works for moving cameras and you can export everything you mentioned by running our algorithm. But in case of dynamic cameras, our rendered background is not that compelling compared to videos with static cameras. You can refer to the Fig.16 of our supplementary document to see more examples of our background rendering.