https://github.com/guoriyue/3dgs-warp-scratch
Build 3D Gaussian Splatting from scratch with NVIDIA Warp in Python — CPU/GPU compatible, with a clean and minimalist design focused on learning modern graphics.
https://github.com/guoriyue/3dgs-warp-scratch
3dgs build-from-scratch cuda graphics nerf nvidia-warp python
Last synced: 4 months ago
JSON representation
Build 3D Gaussian Splatting from scratch with NVIDIA Warp in Python — CPU/GPU compatible, with a clean and minimalist design focused on learning modern graphics.
- Host: GitHub
- URL: https://github.com/guoriyue/3dgs-warp-scratch
- Owner: guoriyue
- License: agpl-3.0
- Created: 2024-06-10T16:12:21.000Z (about 2 years ago)
- Default Branch: main
- Last Pushed: 2025-09-03T07:10:15.000Z (10 months ago)
- Last Synced: 2025-09-03T09:10:52.185Z (10 months ago)
- Topics: 3dgs, build-from-scratch, cuda, graphics, nerf, nvidia-warp, python
- Language: Python
- Homepage:
- Size: 31.1 MB
- Stars: 195
- Watchers: 3
- Forks: 12
- Open Issues: 0
-
Metadata Files:
- Readme: readme.md
- License: LICENSE
Awesome Lists containing this project
- awesome-3D-gaussian-splatting - Warp 3DGS - 3.0 | Warp-based implementation | (Implementations / Community Implementations)
- 3d-resources - Warp 3DGS - based implementation (AGPL-3.0). (Tools, Pipeline & Utilities / Gaussian Splatting & NeRF)
README
# 3D Gaussian Splatting in Python with NVIDIA Warp
This project reimplements the core ideas of 3D Gaussian Splatting in a clean, minimalist Python codebase using NVIDIA Warp. It runs on both CPU and GPU with no CUDA setup, focuses on clarity and parallelism, and is designed as a practical entry point for learning modern graphics and differentiable rendering.
## Why This Implementation?
### ✅ CPU & GPU with Zero Hassle
Thanks to Warp, the same kernel code runs seamlessly on both CPU and GPU — no need to deal with CUDA setup, driver issues, or device-specific kernels. Just flip one config line.
### 🧠 Learn Modern Graphics the Easy Way
Explore differentiable rendering and parallel graphics through clean, readable Python — no pricey GPUs, complex toolchains, or heavy C++/CUDA boilerplate needed.
### 📦 Minimalist & Educational
This isn’t another massive codebase. It’s designed for clarity and experimentation. Strips away complexity so you can focus on understanding how Gaussian Splatting really works.
## Quick Start
### Installation
```bash
# Clone the repository
git clone https://github.com/guoriyue/3dgs-warp-scratch.git
cd 3dgs-warp-scratch
# Install dependencies
pip install warp-lang==1.7.0 numpy==1.26.4 matplotlib==3.9.2 imageio==2.34.1 tqdm==4.66.5 plyfile torch==2.6.0
```
### Download Example Data
```bash
# Download the Lego dataset
bash download_example_data.sh
```
### Rendering
```bash
# Render 3 Gaussian points – a minimalist example
python render.py
```
You should see 3 Gaussian points like:
### Training
```bash
# Train on Lego dataset (CPU by default)
# For GPU training, change DEVICE in config.py to "cuda"
python train.py
```
## Project Structure
```
├── forward.py # 3DGS: Forward pass (reimplementation of graphdeco-inria/gaussian-splatting)
├── backward.py # 3DGS: Backward pass (reimplementation of graphdeco-inria/gaussian-splatting)
├── train.py # Main training loop
├── render.py # Rendering script to validate outputs; confirms forward pass correctness
├── config.py # Configuration and training parameters
├── loss.py # Loss functions for training (includes depth loss, though unused in this repo)
├── scheduler.py # Learning rate scheduler
├── optimizer.py # Adam optimizer and densify & prune logic
├── utils/
│ ├── camera_utils.py # Load camera intrinsics and extrinsics from training data
│ ├── point_cloud_utils.py # Point cloud I/O utilities (e.g., saving to .ply)
│ ├── math_utils.py # General math utilities (e.g., transformation matrices)
│ └── wp_utils.py # Warp utilities for math operations and device transfer
└── data/ # Contains the NeRF-synthetic 'Lego' dataset
```
`forward.py` and `backward.py` are based on [graphdeco-inria/gaussian-splatting](https://github.com/graphdeco-inria/gaussian-splatting). The original pure CUDA version is now reimplemented in Nvidia Warp, easy to understand, set up, and run.
Densification and pruning logic is based on [yzslab/gaussian-splatting-lightning](https://github.com/yzslab/gaussian-splatting-lightning), but restructured here with minimal data preparation and simplified training logic.
## License
This project is licensed under the **GNU Affero General Public License v3.0**.
See the [LICENSE](./LICENSE) file for details.