search

NVlabs / nvdiffrec

Official code for the CVPR 2022 (oral) paper "Extracting Triangular 3D Models, Materials, and Lighting From Images".
Other
2.09k stars 222 forks source link

what is system required #137

Open najafi1350 opened 1 year ago

najafi1350 commented 1 year ago

Hello dear What system is required to run the program? Such as CPU GPU &... thanks

jmunkberg commented 1 year ago

Hello @najafi1350 ,

As stated in the readme: Our approach is designed for high-end NVIDIA GPUs with large amounts of memory. To run on mid-range GPU's, reduce the batch size parameter in the .json files.

The code relies on CUDA, so will only run on NVIDIA GPUs, and I would recommend 16GB or more of GPU memory. For the paper, we used V100 GPUs with 32 GB and A6000 with 48 GB.

We also use the tiny-cuda-nn library, and I refer to their documentation for GPU and cuda version requirements: https://github.com/NVlabs/tiny-cuda-nn

Our code should run fine on both Windows and Linux.