| title | emoji | colorFrom | colorTo | sdk | pinned | license | app_port |
|---|---|---|---|---|---|---|---|
3DGS-Library |
🏢 |
green |
indigo |
docker |
false |
mit |
3000 |
Welcome to the 3DGS-Library, a curated collection of state-of-the-art 3D Gaussian Splatting(3DGS) methods and popular NeRF datasets. This repository aims to provide researchers and practitioners with a one-stop-shop for evaluating and comparing different 3DGS techniques on a diverse set of NeRF datasets.
The 3DGS-Library contains implementations of 5 popular 3DGS methods, including:
- Vanilla 3DGS
- Scaffold-GS
- Mip-Splatting
- GaussianShader
- SuGaR
These methods have been carefully selected to represent the latest advancements in the field of 3D geometry synthesis.
In addition to the 3DGS methods, the library also includes 15 widely-used NeRF datasets, covering a variety of scenes and object types. This extensive dataset collection allows users to thoroughly evaluate the performance of 3DGS models across different challenges and settings.
The primary goal of the 3DGS-Library is to facilitate the comparison and benchmarking of 3DGS techniques, enabling researchers and developers to make informed decisions when selecting the most suitable methods for their specific use cases. By providing a standardized environment and a comprehensive set of resources, we aim to accelerate the progress in novel view synthesis research.
We encourage the community to actively contribute to the 3DGS-Library by submitting new 3DGS methods, or NeRF datasets. Together, we can build a valuable resource that drives innovation and advances the state of the art in novel view synthesis and related fields.
This repository is heavily based on Dylan Ebert's excellent work igf.
This repository is distributed under the MIT License, derived from igf by Dylan Ebert. See the LICENSE.txt file for more information.