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.
[RAL' 25 & IROS‘ 25] MapEval: Towards Unified, Robust and Efficient SLAM Map Evaluation Framework.
| Date | Stars |
|---|---|
| 2026-07-30 | 481 |
| 2026-08-09 | 483 |
| 2026-08-17 | 485 |
| 2026-08-18 | 486 |
| 2026-08-27 | 487 |
| 2026-09-15 | 487 |
| 2026-09-16 | 488 |
| 2026-10-01 | 489 |
| 2026-10-22 | 490 |
| 2026-10-28 | 490 |
| 2026-10-30 | 491 |
| 2026-11-04 | 491 |
Today
— stars today
This week
+1 stars this week
This month
+2 stars this month
Momentum
0.0
growth rate 0.20%/day
<div align="center"> # MapEval: Towards Unified, Robust and Efficient SLAM Map Evaluation Framework [**Xiangcheng Hu**](https://github.com/JokerJohn)<sup>1</sup> · [**Jin Wu**](https://zarathustr.github.io/)<sup>1</sup> · [**Mingkai Jia**](https://github.com/MKJia)<sup>1</sup> · [**Hongyu Yan**](https://scholar.google.com/citations?user=TeKnXhkAAAAJ&hl=zh-CN)<sup>1</sup> · [**Yi Jiang**](https://yijiang1992.github.io/)<sup>2</sup> · [**Binqian Jiang**](https://github.com/lewisjiang/)<sup>1</sup> <br> [**Wei Zhang**](https://ece.hkust.edu.hk/eeweiz)<sup>1</sup> · [**Wei He**](https://sites.google.com/view/drweihecv/home/)<sup>3</sup> · [**Ping Tan**](https://facultyprofiles.hkust.edu.hk/profiles.php?profile=ping-tan-pingtan#publications)<sup>1*†</sup> <sup>1</sup>**HKUST**     <sup>2</sup>**CityU**     <sup>3</sup>**USTB** <br> †Project Lead   *Corresponding Author <a href="https://ieeexplore.ieee.org/document/10910156"><img src='https://img.shields.io/badge/IEEE%20RAL%202025-MapEval-red' alt='Paper PDF'></a>[](https://github.com/JokerJohn/Cloud_Map_Evaluation/stargazers)<a href="https://github.com/JokerJohn/Cloud_Map_Evaluation/network/members"> <img alt="GitHub Forks" src="https://img.shields.io/github/forks/JokerJohn/Cloud_Map_Evaluation?color=white" /> </a><a href="https://github.com/JokerJohn/Cloud_Map_Evaluation/pulls"> <img alt="PRs Welcome" src="https://img.shields.io/badge/PRs-Welcome-white" /> </a>[](https://github.com/JokerJohn/Cloud_Map_Evaluation/issues)[](https://opensource.org/licenses/MIT) </div>  ## Overview MapEval is a comprehensive framework for evaluating point cloud maps in SLAM systems, addressing two fundamentally distinct aspects of map quality assessment: 1. **Global Geometric Accuracy**: Measures the absolute geometric fidelity of the reconstructed map compared to ground truth. This aspect is crucial as SLAM systems often accumulate drift over long trajectories, leading to global deformation. 2. **Local Structural Consistency**: Evaluates the preservation of local geometric features and structural relationships, which is essential for tasks like obstacle avoidance and local planning, even when global accuracy may be compromised. These complementary aspects require different evaluation approaches, as global drift may exist despite excellent local reconstruction, or conversely, good global alignment might mask local inconsistencies. Our framework provides a unified solution through both traditional metrics and novel evaluation methods based on optimal transport theory. ## News - **2025/07/06:** Use TBB to accelerate MME calculation, update parameter settings and add more configuration examples. - **2025/05/05:** Add new test data and remove simulation codes. - **2025/03/05:** [Formally published](https://ieeexplore.ieee.org/document/10910156) in IEEE RAL! - **2025/02/25:** Paper accepted! - **2025/02/12:** Source code released! - **2025/02/05:** Paper resubmitted. - **2024/12/19:** Paper submitted to **IEEE RAL**! ## Key Features ### Traditional Metrics Implementation - **Accuracy (AC)**: Point-level geometric error assessment - **Completeness (COM)**: Map coverage evaluation - **Chamfer Distance (CD)**: Bidirectional point cloud difference - **Mean Map Entropy (MME)**: Information-theoretic local consistency metric ### Novel Proposed Metrics - **Average Wasserstein Distance (AWD)**: Robust global geometric accuracy assessment - **Spatial Consistency Score (SCS)**: Enhanced local consistency evaluation  ## Experimental Results ### Simulation Experiments | Noise Sensitivity
Excerpt of 14,821 characters
Read on GitHubWould you bet a product on this? Bounded 0–100 and slow moving.
matched fp:8dfafa0882c4ac9e, topic:robotics, topic:slam, desc:slam