Dandan Zhu
Papers
1
Total Citations
5
H-Index
1
About
Dandan Zhu is a researcher at the forefront of computer vision, with a primary focus on person detection and recognition in challenging, real-world environments. Her work is distinguished by a deep engagement with geometric deep learning, particularly through the development of group-equivariant neural networks. Her most-cited paper, "GET: group equivariant transformer for person detection of overhead fisheye images" (2023), introduces a novel architecture that explicitly encodes rotational symmetries, enabling robust detection in the highly distorted, panoramic views typical of overhead fisheye cameras. This contribution directly addresses a critical bottleneck in surveillance, autonomous navigation, and crowd analysis, where standard convolutional networks often fail. While her citation count is early-stage, the conceptual impact of her work is significant, laying a foundation for more geometrically-aware and data-efficient models. Zhu’s research elegantly bridges theoretical advances in equivariance with practical, deployment-ready solutions, marking her as a promising voice in the next generation of vision researchers.
Research Focus
Key Achievements
Top Papers
- 1