Max Robotics — Global Service Robot Directory + US Certification

About

Jinkai Li is a leading researcher in robotics and human-robot interaction, with a focus on enabling machines to perceive, predict, and act in dynamic environments. His work spans three key areas: robotic manipulation, human motion prediction, and 3D gaze estimation. Li’s early contributions include a seminal 2006 paper on grasping unknown objects using 3D model reconstruction (50 citations), which laid the groundwork for autonomous robotic handling of unstructured objects. More recently, he has advanced human motion prediction with innovative architectures like the Adaptive Multi-level Hypergraph Convolution Network (AMHGCN, 2024, 21 citations) and the Component-wise Self-Correction Network (2025), which improve the accuracy and realism of predicted human poses. Li has also made strides in social robotics through EasyGaze3D (2023, 5 citations), a flexible method for 3D gaze estimation from a single RGB camera, enabling robots to better interpret human intentions. His work on skill learning in robot-assisted micro-manipulation (2024) further demonstrates his commitment to practical applications, from manufacturing to healthcare. With over 90 total citations and a growing portfolio of high-impact publications, Li is shaping the future of intelligent, human-aware robotic systems.