PIG: Physics-Informed Gaussians as Adaptive Parametric Mesh Representations

Namgyu Kang · Jaemin Oh · Youngjoon Hong · Eunbyung Park

Paper | Project Page

Allen_Cahn_Gaussians.mp4

Klein_Gordon_Gaussians.mp4

Helmholtz_Gaussians.mp4

Quick Start

1. Installation

Clone Physics-Informed-Gaussians repo

git clone https://github.com/NamGyuKang/Physics-Informed-Gaussians.git
cd Physics-Informed-Gaussians

Create JAX environment (Flow-Mixing, Klein-Gordon, Nonliner-Diffusion Eq.)

Please follow the steps in the Jax_gpu_version_installation.txt file to install JAX GPU version.

Create Pytorch environment (Helmholtz Eq.)

The code is tested with Python (3.8, 3.9) and PyTorch (1.11, 11.2) with CUDA (>=11.3). You can create an anaconda environment with those requirements by running:

• if you use CUDA 11.3, Pytorch 1.11, Python 3.9, conda env create -f CUDA_11_3_Pytorch_1_11_Py_3_9.yml
• or with CUDA 11.6, Pytorch 1.12, Python 3.8, conda env create -f CUDA_11_6_Pytorch_1_12_Py_3_8.yml
• and then conda activate pig

2. Run the code in each folder

• CUDA_VISIBLE_DEVICES=0 bash flow_mixing3d_pig.sh
• CUDA_VISIBLE_DEVICES=0 bash helmholtz2d_pig.sh
• CUDA_VISIBLE_DEVICES=0 bash klein_gordon3d_pig.sh
• CUDA_VISIBLE_DEVICES=0 bash diffusion3d_pig.sh

Citation

If you find this code useful in your research, please consider citing us!

@article{kang2024pig,
  title={PIG: Physics-Informed Gaussians as Adaptive Parametric Mesh Representations},
  author={Kang, Namgyu and Oh, Jaemin and Hong, Youngjoon and Park, Eunbyung},
  journal={arXiv preprint arXiv:2412.05994},
  year={2024}
}

Contact

Contact Namgyu Kang if you have any further questions.

Acknowledgements

This project is built on top of several outstanding repositories: SPINN, PIXEL, JAXPI. We thank the original authors for opensourcing their excellent work.