Jingsheng Lei

Zhejiang University of Science and Technology

Papers

Total Citations

350

H-Index

About

Jingsheng Lei is a leading researcher in computer vision, with a primary focus on semantic segmentation for autonomous driving and robotic perception. His work centers on developing advanced deep learning architectures that effectively integrate multi-modal sensor data—specifically RGB, thermal, and depth imagery—to achieve robust scene understanding in complex urban environments. Lei’s major contribution is the creation of novel network designs that address the challenge of fusing cross-modal information, as exemplified by his most cited work, "GMNet: Graded-Feature Multilabel-Learning Network for RGB-Thermal Urban Scene Semantic Segmentation" (2021, 318 citations). This paper introduced a graded-feature multilabel-learning approach that significantly improves segmentation accuracy in low-light or adverse weather conditions by leveraging complementary thermal data. More recently, his "THCANet: Two-layer hop cascaded asymptotic network for robot-driving road-scene semantic segmentation in RGB-D images" (2023, 32 citations) further advances the field by optimizing depth-aware feature extraction for real-time robotic applications. With over 350 total citations and growing influence, Lei’s work is foundational for next-generation autonomous systems, directly impacting safety and reliability in self-driving cars and intelligent robots.