Zhengwen Shen

China University of Mining and Technology

Papers

Total Citations

H-Index

About

Zhengwen Shen is a rising researcher in computer vision and autonomous driving perception, with a focus on multimodal urban scene understanding. His most notable contribution is the development of HEFANet (Hierarchical Efficient Fusion and Aggregation Segmentation Network), a novel architecture for RGB-thermal urban scene parsing. This work addresses the critical challenge of fusing visible and thermal imagery for robust semantic segmentation in adverse conditions, such as low light or fog. By introducing efficient hierarchical fusion and aggregation mechanisms, HEFANet achieves state-of-the-art performance on standard benchmarks, demonstrating significant improvements in parsing accuracy while maintaining computational efficiency. Although recently published in 2024, the paper has already garnered 3 citations, signaling growing interest in his approach. Shen’s research directly impacts autonomous navigation and surveillance systems, where reliable scene parsing under varying illumination is essential. His work exemplifies the trend toward leveraging complementary sensor modalities for more resilient AI perception. As an early-career researcher, Shen is establishing himself at the intersection of efficient deep learning and multimodal fusion, with potential for further breakthroughs in real-time urban scene analysis.