Moritz Kampelmühler

fp16 compatibility

3

Thanks for the great repo! Can you point me to what needs to be done to enable fp16 compatibility? I suppose it will need some rewriting in the CUDA kernels....

bug: remove_infinite_values() can't work for faces

3

`remove_infinite_values()` is using `np.isfinite()`, which will only work for `float` types, but the faces array is `np.int64` which can not represent `np.nan` or `np.inf`. https://github.com/mikedh/trimesh/blob/main/trimesh/base.py#L1246 Since `np.full(1, fill_value=np.inf, dtype=np.int64)` and...

OBJ loading and export causing self-intersections

5

I do have an obj file representing the following geometry  when I load this obj file with trimesh and export it again directly the mesh will have defects. This...

convTri function in custom_functional.py doesn't do anything

5

There is a return before actual execution of the function code, is this expected bahavior? https://github.com/nv-tlabs/GSCNN/blob/3648b86d822f3e73ab1d8e827cf47466bf8f01a7/my_functionals/custom_functional.py#L93

Tutorial 3: missing data

2

All the example data for Tutorial 3 is not in the repository https://github.com/libigl/libigl-python-bindings/blob/40be4c4fa5b8b77d37128b944883656d809d2444/tutorial/tut-chapter3.ipynb

MxNet Compilation flags

2

Hi, for me demo.py wouldn't execute without errors, unless I compiled mxnet with the `USE_CUDA=1 USE_CUDA_PATH=/usr/local/cuda-8.0 USE_CUDNN=1` flags. Maybe you could/should add this to the README. Cheers, Moritz

Hi! Can you please explain why if the training mode is 'generator' the discriminator is trained? Thanks. https://github.com/taesungp/swapping-autoencoder-pytorch/blob/d67e60f8a702868d08e5b943fc7f46908d78e48b/optimizers/swapping_autoencoder_optimizer.py#L61-L64

The warning is triggered because when `force="mesh"` `skip_materials=True` will basically be ignored: https://github.com/mikedh/trimesh/blob/14ff34057ade2a55168fa511c78f721ad8a51e13/trimesh/exchange/load.py#L166 Is this intended behavior?

Bug: Laplacian smoothing not working after hole filling

1

After using `meshing_close_holes()` on a MeshSet, `apply_coord_laplacian_smoothing()` won't have any effect. Details in mwe: ``` import pymeshlab pymeshlab.print_pymeshlab_version() mesh_path ="bunny10k.ply" ms = pymeshlab.MeshSet() ms.load_new_mesh(mesh_path) ms.meshing_close_holes() verts_before = ms.current_mesh().vertex_matrix() ms.apply_coord_laplacian_smoothing() verts_after...