Monte Carlo Energy Aggregation for Mobile 3D Gaussian Splatting

If you like our project, please give us a star ⭐ on GitHub for latest update.

😮 Highlights

🚩 Updates

Welcome to watch 👀 this repository for the latest updates.

✅ [2026.6.30] : You are free to use the ideas of Flux-GS for commercial usage.

✅ [2026.6.30] : The code of our WebGL mobile renderer is realease, you can refer to code.

✅ [2026.6.30] : Release project page.

✅ [2026.6.30] : Code Release.

Setup

For installation: We recommend to use cuda 12.6 with python 3.11 for easy setup.

git clone git@github.com:xiaobiaodu/Flux-GS.git

conda create -n flux-gs python==3.11
conda activate flux-gs

pip install torch torchvision --index-url https://download.pytorch.org/whl/cu126
pip install --no-build-isolation -r requirements.txt

Install TMC (GPCC), and add tmc3 to your environment variable or manually specify its location in the code (lines 243 and 258, this script is sourced from HAC++).

If you have trouble in installing cuml, please refer to the CUML Installation Guide.

We used Mip-NeRF 360, Tanks & Temples, and Deep Blending.

Running

sh train.sh

# To improve rendering perofmrance, you can use multi-view training from MVGS. It may cause longer training time and memory.
python train.py ... --mv  3

Evaluation

The trained weights are released on Huggingface

python render.py -s <path to COLMAP> -m <model path> --decode
python metrics.py -m <model path> 

--decode

Rendering with the compressed file (comp.json), otherwise using the ply file. The results are the same regardless of this option.

Mobile Rendering

We save the trained Gaussian Splatting file as .json for better loading in WebGL with point cloud and MLP decompression. Our WebGL mobile render is open to the public. The code is (here)[https://github.com/xiaobiaodu/flux-gs-project]

👍 Acknowledgement

This work is built on many amazing research works and open-source projects, thanks a lot to all the authors for sharing!

• Mobile-GS
• MVGS
• FastGS

BibTeX

@misc{du2026mobile-gs,
      title={Mobile-GS: Real-time Gaussian Splatting for Mobile Devices}, 
      author={Xiaobiao Du and Yida Wang and Kun Zhan and Xin Yu},
      year={2026},
      eprint={2603.11531},
      archivePrefix={arXiv},
      primaryClass={cs.CV},
      url={https://arxiv.org/abs/2603.11531}, 
}

@inproceedings{du2026fluxgs,
  title={Monte Carlo Energy Aggregation for Mobile 3D Gaussian Splatting},
      author={Xiaobiao Du and Yuan Wang, and Hao Li, and Bosheng wang, and Xun Sun,  and Xin Yu},
  booktitle={European Conference on Computer Vision (ECCV)},
  year={2026}
}