Skip to content
Projects

LioQing/wgpu-3dgs-viewer-app

View on GitHub

LioQing/wgpu-3dgs-viewer-app has 48 GitHub stars and 8 forks. GitHot measured 0 stars in 24 hours and 0 in 7 days. Last checked 2026-11-14 21:09 UTC. Last push 2025-05-16.

How these counts are measured

What you can do

  • Load large PLY files at very fast speed.
  • Different Gaussian display modes.
  • Render multiple models at once.
  • Edit the model's transform and colors.

Measured star growth

in 24h
0
in 7 days
0
in 15 days
Collecting history
in 1 month
Collecting history

Net star changes over overlapping periods, not daily totals. Missing measurements are left blank. Stars measure attention, not software quality.

About

A 3D Gaussian Splatting Viewer App written in Rust using wgpu and egui.

Creative & Media

README.md

Original

3D Gaussian Splatting Viewer App

...written in Rust using wgpu and egui.

Github Pages CI License

Getting started

This viewer app is built for 3D Gaussian Splatting. It supports the PLY file format from the 3D Gaussian Splatting for Real-Time Radiance Field Rendering research paper.

[!NOTE]

The core of this app is a rendering crate for 3D Gaussian splatting, wgpu-3dgs-viewer! The crate uses wgpu, the Rust implementation of WebGPU, to render the model, so it supports most of the GPU backends.

Running the app

Web

You can run the app on the web by visiting https://lioqing.com/wgpu-3dgs-viewer-app/.

[!NOTE]

To use the web version, you need a browser that supports WebGPU. Please refer to the WebGPU Implementation Status for more information.

[!WARNING]

Apple Silicon is found to crash when trying to render models in the web version, use the native version instead.

Native

You can run the app natively, by downloading from the latest release from the releases page depending on your operating system.

Features

  • Load large PLY files at very fast speed.
  • Different Gaussian display modes.
  • Render multiple models at once.
  • Edit the model's transform and colors.
  • Mask the model with composite shapes.

Demo

The following demos are recorded on laptop with:

  • CPU: AMD Ryzen 9 4900HS
  • GPU: NVIDIA GeForce RTX 2060 with Max-Q Design
  • RAM: 32 GB

Compression & Loading

Models are compressed and loaded while being streamed from the file at a very fast speed.

https://github.com/user-attachments/assets/9f3ba3de-e7b9-4a86-baae-98574a93a21f

Display Mode

Gaussians can be viewed as ellipses or points, with specified spherical harmonic degrees, and at different sizes.

https://github.com/user-attachments/assets/bfc36e2a-0939-4703-af47-3c958556770a

Multiple Models

Multiple models can be loaded to be viewed in the same scene.

https://github.com/user-attachments/assets/45e05ed1-ad93-44ea-9213-5218748cd26e

Selection & Editing

Gaussians can be selected using various tools and be used for color editing.

https://github.com/user-attachments/assets/5a2ef198-f8a3-43db-94ff-9e393041bf1e

Composite Mask

Masks can be used to select Gaussians to view, which can be composed using set operations.

https://github.com/user-attachments/assets/3ec2a95f-291b-4684-9023-94a57d8c3b2a

Development

Native

Make sure you are using the latest version of stable rust by running rustup update.

cargo run --release

On Linux you need to first run:

sudo apt-get install libxcb-render0-dev libxcb-shape0-dev libxcb-xfixes0-dev libxkbcommon-dev libssl-dev

On Fedora Rawhide you need to run:

dnf install clang clang-devel clang-tools-extra libxkbcommon-devel pkg-config openssl-devel libxcb-devel gtk3-devel atk fontconfig-devel

Web locally

We compile to WASM and use Trunk to build for web target.

  1. Install the required target with rustup target add wasm32-unknown-unknown.
  2. Install Trunk with cargo install --locked trunk.
  3. Run trunk serve to build and serve on http://127.0.0.1:8080. Trunk will rebuild automatically if you edit the project.
  4. Open http://127.0.0.1:8080/index.html#dev in a browser. See the warning below.

[!NOTE]

assets/sw.js script will try to cache our app, and loads the cached version when it cannot connect to server allowing your app to work offline (like PWA). appending #dev to index.html will skip this caching, allowing us to load the latest builds during development.