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.
📖[IEEE Sensors Journal (JSEN) ] SuperVINS: A Real-Time Visual-Inertial SLAM Framework for Challenging Imaging Conditions (integrated deep learning features)
| Date | Stars |
|---|---|
| 2026-07-24 | 456 |
| 2026-07-25 | 456 |
| 2026-07-28 | 456 |
| 2026-07-30 | 456 |
| 2026-07-31 | 457 |
| 2026-08-06 | 459 |
Today
+2 stars today
This week
+3 stars this week
This month
— stars this month
Momentum
0.0
growth rate 0.66%/day
<!-- * @Author: Hongkun Luo * @Date: 2024-07-24 03:11:30 * @LastEditors: luohongk [email protected] * @Description: * * Hongkun Luo --> <h2 align="center">SuperVINS: A Real-Time Visual-Inertial SLAM Framework<br>for Challenging Imaging Conditions</h2> <h3 align="center"> <a href="https://luohongkun.top/scholar/">Hongkun Luo</a>, <a href="https://yangliu9527.github.io/">Yang Liu</a>, <a href="https://jszy.whu.edu.cn/guochi">Chi Guo</a>, <a href="https://cesi.cumt.edu.cn/info/1101/8625.htm">Zengke Li</a>, <a href="https://gnsscenter.whu.edu.cn/info/1301/1081.htm">Weiwei Song</a> </h3> <p align="center"> <a href="https://luohongkun.top/SuperVINS/">Project Website</a> | <a href="https://ieeexplore.ieee.org/document/10949688">Paper (IEEE Sensors Journal)</a> | <a href="https://arxiv.org/abs/2407.21348">arXiv</a> </p> <p align="center"> <a href="https://github.com/luohongk/SuperVINS"> <img src="https://img.shields.io/badge/VINS-Image_IMU-red" /> </a> <a href="https://cmake.org/"> <img src="https://img.shields.io/badge/C++-14-blue" /> </a> <a href="https://github.com/cvg/LightGlue"> <img src="https://img.shields.io/badge/SuperPoint+LightGlue-red" /> </a> <a href="http://wiki.ros.org/noetic"> <img src="https://img.shields.io/badge/ROS1-Noetic-blue" /> </a> <a href="https://github.com/rmsalinas/DBow3"> <img src="https://img.shields.io/badge/BoW-DBoW3-red" /> </a> <a href="https://www.gnu.org/licenses/gpl-3.0.html"> <img src="https://img.shields.io/badge/License-GPL3.0-yellow.svg" /> </a> <a href="https://www.zhiyuteam.com/"> <img src="https://img.shields.io/badge/Wuhan_University-BRAIN_LAB-green" /> </a> </p> <div align=center><img src="resources/SuperVINS.png" width=100%></div> --- ## 📢 News - **2026.05**: Released **SuperVINS 2.0** — Fixed memory leak bugs, added LightGlue-based loop closure verification (SuperPoint+LightGlue PnP), dense loop-corrected trajectory output (TUM format), and enhanced RViz visualization. - **2025.03**: Published in **IEEE Sensors Journal**. - **2025.03.07**: Released SuperVINS 1.0 with SuperPoint+LightGlue front-end. - **2024.10.06**: Released base code. - **2024.08.07**: Added demo. - **2024.07.31**: Published preprint. [arXiv](https://arxiv.org/abs/2407.21348) --- ## 🎬 Demo <div align=center><img src="resources/SuperVINS_demo.gif" width=100%></div> Full demo video: [resources/video.mp4](resources/video.mp4) --- ## 🔍 Overview SuperVINS is a real-time Visual-Inertial SLAM system built upon [VINS-Fusion](https://github.com/HKUST-Aerial-Robotics/VINS-Fusion), replacing traditional handcrafted features with deep-learning-based feature extraction and matching throughout the entire pipeline: | Module | VINS-Fusion (Original) | SuperVINS | | ---------------------------- | ------------------------- | ------------------------------------------------------ | | **Feature Extraction** | Shi-Tomasi corners | SuperPoint (learned keypoints + descriptors) | | **Feature Matching** | Optical flow (KLT) | LightGlue (learned matcher via ONNX Runtime GPU) | | **Loop Detection** | DBoW2 + BRIEF | DBoW3 + SuperPoint descriptors | | **Loop Verification** | BRIEF descriptor matching | **SuperPoint + LightGlue matching + PnP RANSAC** | ### Why "SuperVINS"? The name honors **SuperPoint** and **VINS-Fusion** — "Super" refers to the SuperPoint descriptor that runs through both front-end tracking and loop closure, while "VINS" acknowledges the visual-inertial fusion backbone. --- ## 📦 Download Datasets ```bash cd ~/catkin_ws/src/SuperVINS chmod +x download_data.sh ./download_data.sh ``` Or manually download [EuRoC MAV Dataset](https://projects.asl.ethz.ch/datasets/doku.php?id=kmavvisu
Excerpt of 12,945 characters
Read on GitHubWould you bet a product on this? Bounded 0–100 and slow moving.
matched fp:ed60746efde3e285, topic:robotics, topic:slam, readme:robotics