Max Robotics — Global Service Robot Directory + US Certification

About

Cong Bai is a leading researcher at the intersection of computer vision and intelligent systems, with key contributions in underwater image enhancement, surgical instrument segmentation, and human-object interaction (HOI) detection. His work addresses critical challenges in visual perception under complex, real-world conditions. Bai’s most influential paper, “Structure-Inferred Bi-level Model for Underwater Image Enhancement” (2022, 19 citations), tackles the pervasive issues of color cast and low visibility in underwater imagery, a vital problem for autonomous underwater robots. He also developed PaI-Net (2021, 16 citations), a modified U-Net architecture that reduces the semantic gap for precise surgical instrument segmentation in minimally invasive surgery. In the domain of smart city systems, Bai advanced multi-object tracking using attention networks (2022, 11 citations). His more recent work, “SGPT: The Secondary Path Guides the Primary Path in Transformers for HOI Detection” (2023, 7 citations), introduces a novel dual-path transformer framework that significantly improves the detection of human-object interactions, with direct applications in behavior analysis and robotic manipulation. Through these diverse yet interconnected contributions, Bai demonstrates a consistent focus on enhancing machine perception in challenging environments, from the depths of the ocean to the operating room and the urban landscape.