README.md

metadata

task_categories:
  - other
tags:
  - gaussian-splatting
  - 3d-compression

Voxel-GS Dataset

This repository contains datasets and resources related to the paper Voxel-GS: Quantized Scaffold Gaussian Splatting Compression with Run-Length Coding.

Voxel-GS proposes a simple yet highly effective framework for compressing substantial Gaussian splatting format point clouds. It achieves competitive performance using a lightweight rate proxy and run-length coding, departing from complex neural entropy models.

Paper: https://huggingface.co/papers/2512.17528 Code: https://github.com/zb12138/VoxelGS

Dataset Structure

The voxelGS/data directory, as described in the associated GitHub repository, contains various 3D scene datasets structured as follows:

voxelGS/data
├── DeepBlending 
    ├── drjohnson
    └── playroom
├── MipNerf360 
    ├── bicycle
    ├── bonsai
    ├── counter
    ├── flowers
    ├── garden
    ├── kitchen
    ├── room
    ├── stump
    └── treehill
├── Nerf_Synthetic 
    ├── chair
    ├── drums
    ├── ficus
    ├── hotdog
    ├── lego
    ├── materials
    ├── mic
    └── ship
└── T2T 
    ├── train
    └── truck

Sample Usage

The following sections provide instructions for setting up the environment and using the Train.py and Coder.py scripts from the GitHub repository to interact with Voxel-GS.

Environment Setup

To set up the environment with cuda 11.7 and python 3.8:

conda install pytorch==1.13.1 torchvision==0.14.1 torchaudio==0.13.1 pytorch-cuda=11.7 -c pytorch -c nvidia
pip install omegaconf,loguru,open3d==0.19.0,opencv-python,plyfile,tensorboard,termcolor,torch_scatter,jaxtyping,einops,lpips
pip install models/submodules/* # from Scaffold-GS

Training

You can train the model using Train.py. After training, stat_log will generate statistical information and save the results in the output folder.

python Train.py

Alternatively, you can download pre-compressed binary files and decompress them directly:

python Coder.py -bin Results/BIN/DeepBlending/playroom/gsbin -eval

Encoding, Decoding, and Evaluation

The Coder.py script can be used for encoding PLY files, decoding binary files, and for visualization and evaluation.

1. Encode a PLY file:

   python Coder.py -ply output/base/Nerf_Synthetic/hotdog/point_cloud/point_cloud_30000.quantized.ply -encode
   

2. Decode a binary (gsbin) file:

   python Coder.py -bin output/base/Nerf_Synthetic/hotdog/gsbin
   

3. Visualize and Evaluate a binary (gsbin) file:

   python Coder.py -bin output/base/Nerf_Synthetic/hotdog/gsbin -show -eval