Yazhe Tang

Papers

Total Citations

H-Index

About

Yazhe Tang’s research focuses on computer vision and robotics, with key contributions in omnidirectional vision, object recognition, and robust feature extraction. Tang’s most cited work, “Parameterized Distortion-Invariant Feature for Robust Tracking in Omnidirectional Vision” (2015, 6 citations), addresses the challenge of nonlinear distortions in central catadioptric omnidirectional images caused by quadratic mirrors. By developing a parameterized distortion-invariant feature, Tang enables robust tracking on severely distorted images where conventional pin-hole models fail—a critical advancement for panoramic robotic vision systems. In another notable study, “A hybrid shape descriptor for object recognition” (2015, 4 citations), Tang proposes an effective shape contour representation that enhances object characterization for robot vision and automation tasks. This hybrid descriptor improves recognition accuracy by capturing meaningful shape information, supporting applications in autonomous navigation and industrial robotics. Though Tang’s citation counts are modest, the work demonstrates foundational contributions to distortion-tolerant vision and shape-based recognition—essential for real-world robotic perception in non-ideal imaging conditions. These studies highlight Tang’s role in advancing practical computer vision for robotics, particularly in handling complex visual distortions and improving object recognition reliability.