Denis Hadjivelichkov

You can bypass the problem by reassigning the index at very little cost, would still be faster if fixed in the C++ code probably. ```python for entry in slic.slic_model.clusters: y,...

> You might have an old installation of nvidia-cuda-toolkit installed via apt-get (like me). To check, run: `apt-cache policy nvidia-cuda-toolkit` > > I suspect you have that package installed and...

The issue is still happening to me. Specifically, it's caused by pydensecrf/eigen.cpp:28:10 Installing cython did not work for me. Does it have to be a specific version? My python is...

@riflemanl Hi, did you figure this out? I am also getting black images.

I had to export the following: ```bash export TCNN_CUDA_ARCHITECTURE=86 export CUDA_HOME="/usr/local/cuda-11.7" export PATH="/usr/local/cuda-11.7/bin:$PATH" export LD_LIBRARY_PATH="/usr/local/cuda-11.7/lib64:$LD_LIBRARY_PATH" ``` before `example.sh` It solved the ``` import BACore ModuleNotFoundError: No module named 'BACore' ```...

I used meshlab to generate a texture.png from the vertex colors. The export to .obj takes care of the rest. Ref: [https://github.com/facebookresearch/pifuhd/issues/8#issuecomment-645173846](https://github.com/facebookresearch/pifuhd/issues/8#issuecomment-645173846)

This is related to [this issue](https://github.com/lucasb-eyer/pydensecrf/issues/78). Although those solutions don't work for me, they might for you.

> Hi @dennisushi > > Have you solved your question? I would also be interested in knowing the response I ended up using BundleTrack and BundleSDF instead.

`grep -ir "update_training_images" thirdparty/instant-ngp` gives no results