SuperSplat 4D is an enhanced version of the SuperSplat Editor, extending support to Dynamic Gaussian Splatting (4D). It is a free and open-source tool for inspecting, editing, optimizing, and publishing both static and dynamic Gaussian Splats. Built on web technologies, it runs directly in your browser.
A live version of this tool is available at: https://supersplat4d.netlify.app/
- Dynamic Gaussian Support: Load, play, and visualize dynamic splats with Time-Radial Basis Functions (TRBF).
- Format Support:
- .sog4d: Highly compressed dynamic splat format using k-means clustering and WebP.
- .ply: Direct loading of dynamic PLY files containing
trbf_center,trbf_scale, andmotionattributes.
- Dynamic Editing:
- Select and delete artifacts or unwanted splats across the entire animation.
- Filter splats based on dynamic opacity.
- Export Capabilities:
- Export edited dynamic scenes back to
.plyor.sog4d. - Selective time-range export (Scheme A: keeping original
trbf_center).
- Export edited dynamic scenes back to
The repository includes a Python script to compress raw dynamic PLY files into the efficient .sog4d format:
python ply_to_sog4d.py --ply input.ply -o output.sog4dThe script can read cfg_args directly from the PLY header if it contains a line like:
comment cfg_args: duration=2.0 start=0.0 fps=30.0 sh_degree=0
To initialize a local development environment, ensure you have Node.js 18 or later installed.
-
Clone and switch branch:
git clone https://github.com/playcanvas/supersplat.git cd supersplat git checkout splat4d -
Install dependencies:
npm install
-
Run development server:
npm run develop
-
Navigate to
http://localhost:3000
Supported languages are located in static/locales. To test a specific language:
http://localhost:3000/?lng=<locale> (e.g., zh, en).
SuperSplat is made possible by the open source community and the original PlayCanvas team.