Chenggang Yan

Hangzhou Dianzi University, Harbin University of Science and Technology

Papers

Total Citations

H-Index

About

Chenggang Yan is a researcher whose work spans the frontiers of computer vision, 3D reconstruction, and robotics. His primary research areas include spherical image object detection, real-time 3D scene reconstruction, and bionic robot design. Yan’s most notable contribution is his work on Gaussian Label Distribution Learning for spherical image object detection (2023, 8 citations), which addresses a critical challenge in applications ranging from virtual reality to autonomous driving by moving beyond traditional \(l_n\)-norms loss functions. He also developed FastFusion (2022, 3 citations), a real-time indoor scene reconstruction method that overcomes the limitations of previous systems by handling fast sensor motion, significantly broadening the applicability of augmented reality and robotics technologies. Earlier in his career, Yan contributed to the parameterized design of mechanical structures for move-in-mud robots (2006, 2 citations), integrating virtual prototyping databases with bionic principles. His work demonstrates a unique ability to bridge theoretical advances in machine learning with practical engineering challenges, making him a versatile and impactful figure in modern robotics and computer vision research.