Tired of managing PyTorch mesh attributes in your deep learning workflows? NVIDIA Kaolin library has introduced the SurfaceMesh class to make it easier to track attributes such as faces, vertices, normals, face_normals, and others associated with surface meshes. It provides consistent import of these attributes for obj, usd, and gltf formats; eases auto-computing attributes like normals; and simplifies indexing and batching. Check out this video for an overview.
For more details, see the following resources:
◻️ Kaolin installation: https://kaolin.readthedocs.io/en/latest/notes/installation.html
◻️ SurfaceMesh documentation: https://kaolin.readthedocs.io/en/latest/modules/kaolin.rep.surface_mesh.html#kaolin-rep-surfacemesh
◻️ SurfaceMesh tutorial: https://github.com/NVIDIAGameWorks/kaolin/blob/master/examples/tutorial/working_with_meshes.ipynb
◻️ GitHub: https://github.com/NVIDIAGameWorks/kaolin
◻️ Kaolin library documentation: https://kaolin.readthedocs.io/en/latest/
graphics 3d deep learning
Add comment