Kunxu Zhao

Papers

Total Citations

H-Index

About

Kunxu Zhao is a researcher advancing the frontiers of human-robot interaction (HRI), with a core focus on enabling robots to perceive, understand, and navigate around humans in a socially intelligent manner. His work bridges computer vision and robotics, particularly through the development of real-time systems for head detection and gaze estimation, which are critical for deciphering human intent. His 2022 paper, "RTHG: Towards Real-Time Head Detection And Gaze Estimation," has garnered 3 citations for its contribution to this sequential task pipeline. Zhao’s major contributions lie in modeling human social spaces for robot navigation. He pioneered the use of an Asymmetric Gaussian Model (2021, 2 citations) to account for the direction of human gaze—a factor often overlooked—preventing robots from intruding on a person's visual space. Furthermore, he has tackled the challenge of robots approaching conversational groups, employing Generative Adversarial Networks (2023, 2 citations) to generate socially acceptable poses that prioritize human comfort alongside safety. By integrating gaze direction and group dynamics into navigation, Zhao’s work is laying the groundwork for robots that are not just collision-free, but genuinely courteous and context-aware companions.