Weihao Gu

Ajinomoto (United States)

Papers

3

Total Citations

212

H-Index

3

About

Weihao Gu is a leading researcher in autonomous navigation and robotics, whose work focuses on advancing place recognition—a critical capability for self-driving vehicles and robots operating in complex, changing environments. Gu’s most significant contribution is the **OverlapTransformer**, a novel transformer-based architecture for LiDAR-based place recognition that is both efficient and rotation-invariant, achieving 196 citations since 2022. This work directly addresses the challenge of yaw-angle invariance, enabling robust loop closure in SLAM and global localization even when a vehicle’s orientation changes. Gu further extended this concept with a refined rotation-invariant variant, and most recently introduced **ModaLink** (2024), a pioneering cross-modal framework that unifies image and point-cloud data for efficient place recognition—a breakthrough for systems that must retrieve visual information from pre-built LiDAR maps. With a growing citation impact and a clear trajectory from single-modal to cross-modal solutions, Gu’s research is shaping the future of robust, real-time localization for autonomous systems.

Research Focus

Key Achievements

3
H-Index
3
Papers
212
Total Citations
71
Avg Citations/Paper
🏆 Most Cited Paper
OverlapTransformer: An Efficient and Yaw-Angle-Invariant Transformer Network for LiDAR-Based Place Recognition
196 citations · 2022
📈 Most Prolific Year: 2022 (2 Papers)
🤝 Key Collaborators: 14
🏛 Institutions: Ajinomoto (United States)

Top Papers

  1. 1
  2. 2
  3. 3

Key Collaborators

Contact & Links

Available for collaboration
Content generated · 13 days ago