Yuan-Ming Li

Papers

Total Citations

H-Index

About

Yuan-Ming Li is a rising researcher in robotics and computer vision, whose work focuses on enabling robots to understand and interact with the physical world in more intelligent, context-aware ways. His primary research areas include task-oriented robotic grasping, procedural action understanding, and error detection in automated systems. Li’s most notable contribution is his work on **task-oriented 6-DoF grasp pose detection in cluttered environments**, where he addresses a fundamental challenge: humans grasp objects differently depending on the intended task—for example, gripping a knife by the handle to cut versus by the blade to hand it over. His approach moves beyond generic grasp detection to consider the functional context of the object, a critical step toward more dexterous and useful robotic manipulation. Additionally, Li has advanced **error detection in procedural tasks**, modeling multiple normal action representations to identify mistakes in real-world activities, a key capability for AR-assisted guidance and robotic process monitoring. Though early in his career, his work has already garnered attention, with his top-cited paper accumulating 4 citations since 2025. Li’s research sits at the intersection of perception, reasoning, and action, promising to make robots more capable assistants in dynamic, human-centric environments.