Tianshui Chen

Papers

Total Citations

H-Index

About

Tianshui Chen is a researcher working at the intersection of computer vision and intelligent robotics, with a focus on developing practical deep learning solutions for real-world perception challenges. His work spans two primary areas: facial expression recognition and lightweight object detection, both driven by the goal of enabling smarter, more capable robotic systems. In facial expression recognition, Chen has made notable contributions through his 2021 paper on AU-Expression Knowledge Constrained Representation Learning, which addresses a persistent gap between lab-controlled performance and real-world deployment. By incorporating action unit knowledge as a constraint, his approach helps models generalize more effectively to unconstrained environments — a critical step toward emotionally intelligent robotics. This work has accumulated 24 citations, reflecting its relevance to the human-robot interaction community. His 2022 work on Scale-Aware Squeeze-and-Excitation for lightweight object detection tackles the challenge of enabling robots to perceive their surroundings under strict computational constraints. Building on high-resolution network architectures, Chen's method enhances feature representation efficiency without sacrificing detection accuracy, garnering 15 citations. Together, these contributions position Chen as a researcher dedicated to bridging the gap between cutting-edge deep learning and the practical demands of intelligent robotic perception.