Progressive Multi-Scale Light Field Networks

Codebase for Progressive Multi-Scale Light Field Networks (3DV 2022).

Getting Started

1. Download our datasets and extract them to datasets directory.
2. Setup a PyTorch environment and install requirements.txt.
3. To train, run python app.py -c configs/run_mipnet_jon.txt.
   Alternatively, download our trained LFNs to runs.

Interactive Viewer

To use the viewer on Ubuntu, run the following:

sudo apt install libmesa-dev libglfw3
# Required to install pycuda with OpenGL support
echo "CUDA_ENABLE_GL = True" > ~/.aksetup-defaults.py
pip install pycuda glumpy pyopengl
rm ~/.aksetup-defaults.py

python app.py -c configs/run_mipnet_jon.txt --script-mode=viewer

If you get CUBLAS_STATUS_EXECUTION_FAILED while opening the viewer, try running with CUBLAS_WORKSPACE_CONFIG=:0:0. (PyTorch Issue).

Citation

@inproceedings{li2022progressive,
  author={Li, David and Varshney, Amitabh},
  booktitle={2022 International Conference on 3D Vision (3DV)}, 
  title={Progressive Multi-Scale Light Field Networks}, 
  year={2022},
  volume={},
  number={},
  pages={231-241},
  doi={10.1109/3DV57658.2022.00035}}
}

Acknowledgments

• utils/nerf_utils.py is borrowed from krrish94/nerf-pytorch.