Lubin Weng

Chinese Academy of Sciences, Institute of Automation

Papers

Total Citations

116

H-Index

About

Lubin Weng is a leading researcher in computer vision and robotics, whose work bridges the critical gap between perception and autonomous navigation. His most impactful contribution is in the challenging domain of glass-like object segmentation. In his seminal 2021 paper, "Enhanced Boundary Learning for Glass-like Object Segmentation," which has garnered over 100 citations, Weng tackled the notoriously difficult problem of enabling machines to detect transparent surfaces like windows and bottles. This work is pivotal for real-world applications such as robot navigation and grasping, where misidentifying glass can lead to catastrophic failures. By focusing on enhanced boundary learning, he developed methods that allow robots to perceive and interact with transparent obstacles safely. Earlier, Weng laid foundational work in autonomous mobility with his 2010 study on "Vision heading navigation based on navigation curve," which introduced a novel method for correcting robot heading angles using salient feature points. This research demonstrates his long-standing commitment to solving practical navigation challenges, establishing him as a key figure whose work directly advances the reliability of autonomous systems in complex, unstructured environments.