3D Gaussian Splatting as Markov Chain Monte Carlo (Packaged Python Version)

This repository contains the refactored Python code for 3dgs-mcmc. It is forked from commit 7b4fc9f76a1c7b775f69603cb96e70f80c7e6d13. The original code has been refactored to follow the standard Python package structure, while maintaining the same algorithms as the original version.

Features

• Code organized as a standard Python package
• Markov Chain Monte Carlo trainer for 3D Gaussian Splatting
• Integration with reduced-3dgs

Prerequisites

• Pytorch (v2.4 or higher recommended)
• CUDA Toolkit (12.4 recommended, should match with PyTorch version)
• (Optional) cuML for faster vector quantization

(Optional) If you have trouble with gaussian-splatting and reduced-3dgs, try to install it from source:

pip install wheel setuptools
pip install --upgrade git+https://github.com/yindaheng98/gaussian-splatting.git@master --no-build-isolation
pip install --upgrade git+https://github.com/yindaheng98/reduced-3dgs.git@main --no-build-isolation

PyPI Install

pip install --upgrade gaussian-splatting-mcmc

or build latest from source:

pip install wheel setuptools
pip install --upgrade git+https://github.com/yindaheng98/3dgs-mcmc.git@main --no-build-isolation

Development Install

git clone --recursive https://github.com/yindaheng98/reduced-3dgs
cd 3dgs-mcmc
pip install --target . --upgrade --no-deps .

Quick Start

1. Download dataset (T&T+DB COLMAP dataset, size 650MB):

wget https://repo-sam.inria.fr/fungraph/3d-gaussian-splatting/datasets/input/tandt_db.zip -P ./data
unzip data/tandt_db.zip -d data/

2. Train 3DGS-MCMC:

python -m gaussian_splatting_mcmc.train -s data/truck -d output/truck -i 30000 --mode base

3. Render:

python -m gaussian_splatting.render -s data/truck -d output/truck -i 30000 --load_camera output/truck/cameras.json

** NeurIPS 2024 SPOTLIGHT **

3D Gaussian Splatting as Markov Chain Monte Carlo

@inproceedings{kheradmand20243d,
    title = {3D Gaussian Splatting as Markov Chain Monte Carlo},
    author = {Kheradmand, Shakiba and Rebain, Daniel and Sharma, Gopal and Sun, Weiwei and Tseng, Yang-Che and Isack, Hossam and Kar, Abhishek and Tagliasacchi, Andrea and Yi, Kwang Moo},
    booktitle = {Advances in Neural Information Processing Systems (NeurIPS)},
    year = {2024},
    note = {Spotlight Presentation},
   }