Jeremy Reizenstein

@TimZaman If you are happy with the CLA, please accept it and then we can merge this. Otherwise feel free to close this PR.

Do you want a differentiable rendering or just a visualization / forward pass? Do you want to only see the wireframe without the mesh, or both the mesh and the...

Plotly has a facility for viewing a wireframe, and we already have encapsulated it in PyTorch3D for visualizing a PyTorch3D mesh. I think it would be quite easy to add...

Have you measured the speed up? In the applications we care about, saving GPU compute and memory is usually more important than saving Python objects. Most of the time the...

In the latest release 0.4.0 of PyTorch3D, we still had a pure python implementation, and this was slower. See for example #252. The C++ code to do the topology calculation...

> mesh_normal_consistency is shape-dependent rather than topology-dependent. It computes the angle between the normals of neighboring faces and those angles could change when deforming a mesh. It definitely does depend...

This isn't provided in PyTorch3D.

> Can I optimize textures with TexturesUV with given image , geometry, and camera pose, then save the result as obj file so that I can view the textured mesh...

I don't know xatlas and I can't help with your modelling question. Rendering is differentiable with respect to `maps`.

For ico_sphere, the level actually affects the shape produced. There is a reason why a higher level than 0 is useful. For this plane, the level doesn't affect the shape...