Top AI Repos — open-source AI, indexed and scored
Top AI Repos tracks AI repositories on GitHub and answers two different questions about each one: is it moving right now, and would you bet a product on it.
Top AI Repos tracks AI repositories on GitHub and answers two different questions about each one: is it moving right now, and would you bet a product on it.
[3DV 2025, Oral] LoopSplat: Loop Closure by Registering 3D Gaussian Splats
| Date | Stars |
|---|---|
| 2026-07-24 | 392 |
| 2026-07-25 | 392 |
| 2026-07-28 | 392 |
| 2026-07-30 | 392 |
| 2026-08-06 | 392 |
Today
— stars today
This week
— stars this week
This month
— stars this month
Momentum
0.0
growth rate 0.00%/day
<p align="center">
<h2 align="center">LoopSplat: Loop Closure by Registering 3D Gaussian Splats</h2>
<p align="center">
<a href="https://www.zhuliyuan.net/"><strong>Liyuan Zhu</strong></a><sup>1</sup>
·
<a href="https://unique1i.github.io/"><strong>Yue Li</strong></a><sup>2</sup>
·
<a href="https://scholar.google.com/citations?user=phiETm4AAAAJ&hl=en/"><strong>Erik Sandström</strong></a><sup>3</sup>
·
<a href="https://shengyuh.github.io/"><strong>Shengyu Huang</strong></a><sup>3</sup>
·
<a href="https://prs.igp.ethz.ch/group/people/person-detail.schindler.html/"><strong>Konrad Schindler</strong></a><sup>3</sup>
·
<a href="https://ir0.github.io/"><strong>Iro Armeni</strong></a><sup>1</sup>
</p>
<p align="center"><strong>International Conference on 3D Vision (3DV) 2025</strong></a>
<p align="center">
<sup>1</sup>Stanford University · <sup>2</sup>University of Amsterdam · <sup>3</sup>ETH Zurich
</p>
<h3 align="center">
[](https://arxiv.org/abs/2408.10154) [](https://loopsplat.github.io/) [](https://opensource.org/licenses/MIT)
<div align="center"></div>
</p>
<p align="center">
<a href="">
<img src="./assets/loopsplat.gif" width="100%">
</a>
</p>
## 📃 Description
<p align="center">
<a href="">
<img src="./assets/architecture.png" width="100%">
</a>
</p>
**LoopSplat** is a coupled RGB-D SLAM system that uses Gaussian splats as a unified scene representation for tracking, mapping, and maintaining global consistency. In the front-end, it continuously estimates the camera position while constructing the scene using Gaussian splats submaps. When the camera traverses beyond a predefined threshold, the current submap is finalized, and a new one is initiated. Concurrently, the back-end loop closure module monitors for location revisits. Upon detecting a loop, the system generates a pose graph, incorporating loop edge constraints derived from our proposed 3DGS registration. Subsequently, pose graph optimization (PGO) is executed to refine both camera poses and submaps, ensuring overall spatial coherence.
# 🛠️ Setup
The code has been tested on:
- Ubuntu 22.04 LTS, Python 3.10.14, CUDA 12.2, GeForce RTX 4090/RTX 3090
- CentOS Linux 7, Python 3.12.1, CUDA 12.4, A100/A6000
## 📦 Repository
Clone the repo with `--recursive` because we have submodules:
```
git clone --recursive [email protected]:GradientSpaces/LoopSplat.git
cd LoopSplat
```
## 💻 Installation
Make sure that gcc and g++ paths on your system are exported:
```
export CC=<gcc path>
export CXX=<g++ path>
```
To find the <i>gcc path</i> and <i>g++ path</i> on your machine you can use <i>which gcc</i>.
Then setup environment from the provided conda environment file,
```
conda create -n loop_splat -c nvidia/label/cuda-12.1.0 cuda=12.1 cuda-toolkit=12.1 cuda-nvcc=12.1
conda env update --file environment.yml --prune
conda activate loop_splat
pip install -r requirements.txt
```
You will also need to install <i>hloc</i> for loop detection and 3DGS registration.
```
cd thirdparty/Hierarchical-Localization
python -m pip install -e .
cd ../..
```
We tested our code on RTX4090 and RTX A6000 GPUs respectively and Ubuntu22 and CentOS7.5.
## 🚀 Usage
Here we elaborate on how to load the necessary data, configure Gaussian-SLAM for your use-case, debug it, and how to reproduce the results mentioned in the paper.
<!-- <details>
<summary><b>Downloading the Data</b></summary> -->
### Downloading the Datasets
We tested our code on Replica, TUM_RGBD, ScanNet, and ScanNet++ datasets. We also provide scripts for downloading Replica and TUM_RGBD in `scripts` folder. Install git lfs before using the scripts by running ```git lfs install```.
For reconstruction evaluation on <b>ReplicaExcerpt of 6,714 characters
Read on GitHubWould you bet a product on this? Bounded 0–100 and slow moving.
matched fp:24a2ebe680d22f7a, topic:slam, readme:slam
matched fp:24a2ebe680d22f7a, topic:gaussian-splatting