Hsin-Ying Lee
Papers
2
Total Citations
27
H-Index
2
About
Hsin-Ying Lee is a rising robotics researcher whose work bridges computer vision and manipulation, with a focus on enabling precise, real-world robotic assembly. Lee’s key research areas include visual servoing, point cloud registration, and 6-degree-of-freedom (6-DoF) object manipulation. Their most notable contribution is the Coarse-to-Fine Visual Servoing (CFVS) framework, which tackles the notoriously difficult peg-in-hole assembly task. By breaking the problem into coarse and fine stages, CFVS achieves high accuracy without restricting the robot’s degrees of freedom or requiring the target to be nearby—a significant leap over prior methods. This work has already garnered 14 citations since its 2023 publication. Complementing this, Lee developed a coarse-to-fine point cloud registration method using SE(3)-equivariant representations, which robustly handles partial overlaps and large pose differences, earning 13 citations in the same year. Together, these contributions demonstrate a systematic approach to overcoming the limitations of both local feature matching and global shape alignment. Lee’s work is particularly impactful for industrial automation and service robotics, where high-precision assembly remains a bottleneck. Their research is a must-read for students and engineers interested in marrying geometric deep learning with practical manipulation.
Research Focus
Key Achievements
Top Papers
- 1
- 2