You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
The readme says that "You can skip this step if you only want to use RVT-2 as it uses our custom Point-Renderer for rendering. PyTorch3D is required for RVT." However, I found that it is still used (for example here). Is it possible to bypass pytorch3d for RVT-2?
The text was updated successfully, but these errors were encountered:
That's a good point. The augmentation function still uses some utils from PyTorch3D, so the current code bypasses only its rendering functionality. If you want to bypass it altogether, the util functions could be replaced by similar ones from Scipy (more info here). I would be happy to merge any updates if you create one.
The readme says that "You can skip this step if you only want to use RVT-2 as it uses our custom Point-Renderer for rendering. PyTorch3D is required for RVT." However, I found that it is still used (for example here). Is it possible to bypass pytorch3d for RVT-2?
The text was updated successfully, but these errors were encountered: