Max Robotics — Global Service Robot Directory + US Certification

About

Shu Zhang is a computer vision and robotics researcher whose work spans depth estimation, underwater perception, human-computer interaction, and simultaneous localization and mapping (SLAM). With a focus on bridging deep learning methodologies and real-world robotic applications, Zhang has made meaningful contributions to several technically demanding domains. Zhang's most influential work, a two-stage deep regression framework for monocular depth estimation (2020, 15 citations), advances autonomous driving and augmented reality by overcoming the practical limitations of hardware-based depth sensors. Complementing this, Zhang has tackled the particularly challenging domain of underwater computer vision, developing a YOLOv3-based marker detection system optimized for underwater environments (2021, 9 citations) and contributing UWStereo, a large-scale synthetic dataset enabling stereo matching research beneath the surface (2025, 5 citations). Zhang's portfolio also demonstrates a commitment to human-robot interaction, with skeleton-guided gesture recognition work addressing conflict-free control systems using Kinect analysis (2020, 7 citations). More recent research explores dynamic-scene SLAM and novel inertial signal reconstruction using Kolmogorov-Arnold Networks, reflecting a continuously evolving research agenda. Collectively, Zhang's work addresses critical perception challenges facing next-generation intelligent robotic systems.