Nanyu Li
Papers
1
Total Citations
5
H-Index
1
About
Nanyu Li is a researcher specializing in computer vision and deep learning, with a particular focus on person detection in challenging visual environments. Their most notable contribution is the development of the Group Equivariant Transformer (GET), a novel architecture designed to address the unique geometric distortions present in overhead fisheye images. This work, published in 2023, has already garnered 5 citations, signaling its early impact on the field. By integrating group equivariance into transformer models, Li has advanced the ability of AI systems to robustly detect individuals in wide-angle, surveillance, and autonomous navigation contexts—applications where traditional convolutional approaches often fail. This innovation not only improves accuracy but also enhances the efficiency of learning from limited data, a critical advantage in real-world deployment. Li’s research sits at the intersection of geometric deep learning and practical computer vision, offering solutions that are both theoretically grounded and immediately applicable. As the demand for intelligent monitoring and spatial awareness grows, Li’s work on equivariant representations is poised to influence future developments in human-centric AI systems.
Research Focus
Key Achievements
Top Papers
- 1