Walter Wu

I was confused by this for a while, too. I think the reason behind it is that coarse samples tend to be sparsely located in the space and be far...

> "pip install jax==0.2.20" may solve your problem Thx! Installing jax==0.2.20 successfully resolves the `flax.errors.InvalidRngError`. Sadly, it now causes segmentation fault raised by line 330 of `train.py`. I'm running on...

> I think that jaxlib version may cause the segfault. I suffered same issue. > > "pip install jaxlib==0.1.71+cuda111 -f https://storage.googleapis.com/jax-releases/jax_releases.html" may solve your problem Thx a lot! The issue...

Thank you for sharing the code and the dataset! May I ask if you can also share the GT geometry used to compute chamfer distance on NeRF synthetic dataset? I'm...

> @OverwriteDev I've almost got it working using the EG3D + Deep3DFaceRecon code but it's slightly off I have actually tried using Deep3DFaceRecon to estimate the pose, however, the result...

In my case I had to install PyTorch 1.7.1 from pip -- installing from conda doesn't work for some reason. `pip install torch==1.7.1+cu110 torchvision==0.8.2+cu110 torchaudio==0.7.2 -f https://download.pytorch.org/whl/torch_stable.html`

> python3 setup.py install This worked for me