hsplat

An open-source library for computer-generated holography using primitives.

Associated paper

Gaussian wave splatting for computer-generated holography | SIGGRAPH 2025

Project Page | Paper

Suyeon Choi*, Brian Chao*, Jacqueline Yang, Manu Gopakumar, Gordon Wetzstein
*denotes equal contribution

Getting Started

1. Environment Setup

First, ensure you have a Python environment with PyTorch installed. Then, install the required Python dependencies:

pip install tyro imageio matplotlib tqdm trimesh scipy

Note: open3d and pytorch3d are also required but may need specific installation steps depending on your OS and Python version (e.g., via Conda or pre-built wheels).

2. Install gsplat

Load the submodules in the gsplat folder:

git submodule update --init --recursive

Then install the gsplat library:

# 在项目根目录下执行
cd D:\paper_code\hsplat-main

# 强制用官方预编译版本覆盖本地任何安装痕迹
pip uninstall -y gsplat
pip install -e . --no-build-isolation

3. Data Preparation

Create the data directories:

mkdir -p hsplat/data
mkdir -p hsplat/models

Download Mip-NeRF 360 and NeRF synthetic dataset and place them in the hsplat/data folder.

Using gsplat, you can generate new scenes and store Gaussians in the hsplat/models folder, but we also provide pre-optimized Gaussians so that you can run the code out of the box. Download them from here and place them in the hsplat/models folder.

Overview

The code is organized as follows:

• The dsplat folder contains the phase encoding function (e.g., DPAC) for the SLMs, from the complex-valued wavefront output of hsplat.
• The gsplat folder contains the gsplat library.
• The hsplat folder contains CGH algorithm implementations using primitives.

Running the Code

To run our Gaussian Wave Splatting algorithm, run:

bash scripts/main_gws_light.sh

To run CGH algorithms for other primitives, run

bash scripts/main_pc.sh  # point cloud
bash scripts/main_meshes.sh  # polygon-based CGH
bash scripts/main_gws.sh  # GWS matcinhg number of primitives

Citation

If you find our work useful in your research, please cite:

@article{choi2025gaussian,
  title={Gaussian wave splatting for computer-generated holography},
  author={Choi, Suyeon and Chao, Brian and Yang, Jacqueline and Gopakumar, Manu and Wetzstein, Gordon},
  journal={ACM Transactions on Graphics (TOG)},
  volume={44},
  number={4},
  pages={1--13},
  year={2025},
  publisher={ACM New York, NY, USA}
}

Contact

If you have any questions, please feel free to email Suyeon Choi and Brian Chao.