Zehang Lin
Papers
1
Total Citations
39
H-Index
1
About
Zehang Lin is a leading researcher in computer vision and robotics, with a focus on perception for manipulation in challenging environments. His work addresses the critical problem of enabling robots to interact with transparent and reflective objects—materials that confound standard depth sensors due to their unique visual properties. Lin’s major contribution is the development of **DepthGrasp**, a self-attentive adversarial network with spectral residual that completes missing depth data for transparent objects. This approach overcomes the limitations of prior linear geometric methods, achieving robust depth prediction even on highly refractive surfaces. His research has garnered significant attention, with his most-cited paper accumulating 39 citations since 2021, reflecting its impact on the grasping and manipulation community. By bridging the gap between perception and action for difficult materials, Lin’s work has practical implications for industrial automation, service robotics, and assistive technologies. His innovative use of adversarial learning and attention mechanisms marks a notable achievement in advancing robotic dexterity.
Research Focus
Key Achievements
Top Papers
- 1