Ziyu Zhu

Papers

Total Citations

H-Index

About

Ziyu Zhu is a rising researcher in computer vision and robotics, whose work centers on visual place recognition (VPR) and multi-modal perception for autonomous systems. His most notable contribution, "EFormer-VPR: Fusing Events and Frames with Transformer for Visual Place Recognition" (2024), tackles a critical challenge in mobile robotics: reliable place recognition under adverse conditions like glare or high-speed motion, where traditional cameras fail due to image blurring. By fusing event camera data—which captures per-pixel brightness changes asynchronously—with conventional frames using a Transformer architecture, Zhu’s approach achieves robust performance where prior methods struggle. This pioneering fusion of event-based and frame-based sensing has already garnered early citations, signaling its impact on the field. Zhu’s work is particularly relevant for autonomous driving and drone navigation, where robustness to lighting and motion extremes is paramount. As an emerging scholar, his research bridges the gap between neuromorphic vision and practical robotics, promising to enhance the reliability of visual localization in real-world deployments.