Wai Ho Li

About

Wai Ho Li is a robotics and computer vision researcher whose work bridges the gap between robotic perception and real-world interaction. His core research areas include visual symmetry detection, active robotic manipulation, and assistive technologies for bionic vision. Li’s most significant contributions lie in developing fast, practical algorithms for detecting reflectional symmetry in images—a critical capability for robotic grasping and visual tracking in domestic environments. His 2005 paper on global reflectional symmetry detection (26 citations) introduced a novel gradient-based method that remains influential. Li pioneered the concept of “active perception” through physical object manipulation, as demonstrated in his 2011 work (23 citations) where robots physically nudge objects to achieve robust segmentation and modeling—a departure from passive camera-based systems. In a remarkable interdisciplinary leap, Li applied robotic sensing principles to improve bionic vision, showing how low-resolution visual prostheses could be enhanced with depth and motion data (23 citations). He also developed the eBug open robotics platform (2011), an accessible teaching and research tool. With over 120 total citations, Li’s work demonstrates how fundamental computer vision techniques can be translated into practical robotic systems and life-changing medical technologies.

Research Focus

Key Achievements

Top Papers

1. 1
   Fast global reflectional symmetry detection for robotic grasping and visual tracking

   26 citations · 2005
2. 2
   Segmentation and modeling of visually symmetric objects by robot actions

   23 citations · 2011
3. 3
   Transformative Reality: Improving bionic vision with robotic sensing

   23 citations · 2012
4. 4
   Interactive learning of visually symmetric objects

   12 citations · 2009
5. 5
   eBug - An open robotics platform for teaching and research

   9 citations · 2011
6. 6
   Bilateral Symmetry Detection for Real-time Robotics Applications

   9 citations · 2008
7. 7
   Autonomous segmentation of Near-Symmetric objects through vision and robotic nudging

   8 citations · 2008
8. 8
   Going beyond vision to improve bionic vision

   7 citations · 2013
9. 9
   A Fast Method For Computing Principal Curvatures From Range Images

   3 citations · 2017

Key Collaborators

Contact & Links