Max Robotics — Global Service Robot Directory + US Certification

About

Baojie Fan is a researcher specializing in 3D computer vision, robotic perception, and intelligent manufacturing systems. His work spans object detection, pose estimation, and single object tracking in point clouds, with significant applications in autonomous driving, robotics, and industrial automation. Fan's early contributions focused on real-time vision systems for teleoperation, developing robust frameworks for object tracking and pose estimation under challenging conditions such as large time delays and cluttered backgrounds. His 2018 work on speedup 3D texture-less object recognition for intelligent manufacturing — now approaching 50 citations — addressed the particularly difficult problem of detecting metal parts with featureless surfaces, a critical challenge in robotic assembly and bin-picking scenarios. More recently, Fan has made notable advances in LiDAR-based 3D perception. His HCPVF framework (2023, 30 citations) introduced hierarchical cascaded point-voxel fusion to balance detection accuracy and computational efficiency, while his work on 3D single object tracking has explored transformer architectures, voting-based refinement strategies, and local-to-global feature integration to improve tracking robustness in autonomous driving contexts. Branching into physical robotics, Fan has also contributed to cable-detecting robot design for large bridge inspection. Collectively, his research demonstrates a versatile and impactful career bridging computer vision, deep learning, and real-world robotic systems.