nr3d_lib icon indicating copy to clipboard operation
nr3d_lib copied to clipboard

Modules, operators and utilities for 3D neural rendering in single-object, multi-object, categorical and large-scale scenes.

Results 18 nr3d_lib issues
Sort by recently updated
recently updated
newest added

First of all, thanks for your great work! I have a small question about the installation on 4090, where I get the compute_capability of "89". I have a cuda-toolkit of...

Hi, I have a qustion about git --recurse-submodules. Cloning into 'nr3d_lib'... remote: Enumerating objects: 791, done. remote: Counting objects: 100% (791/791), done. remote: Compressing objects: 100% (574/574), done. remote: Total...

I wonder how long does it take to compile and install the module after executing `pip install -v .`, since it takes really a long time when compiling lotd

Hello Jianfei, I'm genuinely impressed by the remarkable work. But as I attempt to replicate the Streetsurf, it appears that installing the nr3d_lib is mandatory. The build process has already...

thanks for great work! I followed the readme instructions to clone a GitHub project on my server, created a virtual environment, installed the correct versions of torch and pytorch-scatter, and...

`torch.no_grad` needs to be instantiated with parenthesis to function as a decorator (also works for PyTorch 2.1 and later)

Hi~ it seems that i cannot install nr3d_lib by simply typing "pip install -v .", it comes with errors like this "ERROR: Could not build wheels for nr3d-lib, which is...

Thanks for the awesome job! Currently nr3d_lib supports float64, float32 and float16, I think bfloat16 support can be introduced without too much effort and for my experience it can stabilize...

Hi, Thanks your great works!! I am so curious for this problem. Because ray marching is performed on occupancy grid, and there may be situations where the intervals are discontinuous...

Hi, thank you for your excellent work. I encountered this problem when using LoTDEncoding as a positional encoder for my own model when optimizing the scene representation with depth loss...