Dynamic SLAM by Combining Rigid Feature Point Set Modeling and YOLO
Pengchao Ding, Weidong Wang, Xian Wu, Kangle Xu, Dongmei Wu, Zhijiang Du · 2025
Read Dynamic SLAM by Combining Rigid Feature Point Set Modeling and YOLO by Pengchao Ding, Weidong Wang, Xian Wu, Kangle Xu, Dongmei Wu, Zhijiang Du (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/pmc-12788295.txt.
Description
Pengchao Ding and co-authors propose a visual-inertial SLAM system that can locate a device in places where things move. Moving objects can fool standard SLAM, so the system must spot them and judge their motion in real time. It joins YOLO-V5 segmentation with depth threshold methods at pixel level.
To handle blocked views, object depth is drawn out with K-means clustering, and Kalman filtering matches and tracks objects. A factor graph treats rigid and non-rigid objects by class, and an adaptive rigid point set adds more firm targets. Tests on public and self-built data sets check how well the system handles dynamic scenes.
Catalog record
| CREATOR | Pengchao Ding, Weidong Wang, Xian Wu, Kangle Xu, Dongmei Wu, Zhijiang Du |
|---|---|
| YEAR | 2025 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 45 KB |
| VERSION | 1.0 |
| LICENSE | CC BY 4.0 |
| RIGHTS | CC BY |
| SOURCE | Europe PMC |