Xinggang Hu
Papers
2
Total Citations
31
H-Index
2
About
Xinggang Hu is a leading researcher in robotic perception and semantic simultaneous localization and mapping (SLAM), with a focus on enabling mobile robots to intelligently understand and interact with dynamic indoor environments. His work bridges the gap between low-level geometric mapping and high-level object understanding, addressing fundamental challenges in robot autonomy. Hu’s most cited paper (2022, 23 citations) introduces an object-aware SLAM system featuring an efficient quadric initialization method and joint data association, allowing robots to robustly model objects as ellipsoids rather than simple points. This innovation significantly improves semantic mapping accuracy and object-level interaction. His follow-up work (2022, 8 citations) tackles the critical problem of relocalization under changing conditions—such as lighting and viewpoint shifts—by proposing an object-plane co-represented semantic descriptor with graph propagation. This approach outperforms traditional appearance-sensitive methods and ambiguous landmark-based techniques. Hu’s contributions are shaping the next generation of robust, semantically aware robotic systems, making him a key figure in advancing SLAM for real-world deployment.
Research Focus
Key Achievements
Top Papers
- 1
- 2