Hongzi Zhu

Shanghai Jiao Tong University

Papers

6

Total Citations

72

H-Index

4

About

Hongzi Zhu is a researcher whose work bridges computer vision and robotics, with a particular focus on monocular 3D object detection and RFID-based localization for mobile robotic systems. His research addresses the practical challenges of deploying perception algorithms on resource-constrained platforms such as vehicles, drones, and autonomous robots. Among his most influential contributions is MonoATT, an online monocular 3D object detection framework that introduces an Adaptive Token Transformer to overcome the limitations of traditional grid-based vision tokens in low-compute environments, earning 37 citations since 2023. Complementing this, his MoGDE and ground depth estimation works tackle the persistent near-far disparity problem inherent to monocular cameras in dynamic mobile settings. Together, these contributions represent a coherent research agenda aimed at making 3D perception both accurate and computationally feasible. Zhu has also made notable strides in IoT-integrated robotics, developing POLO and subsequent portable RFID localization systems that enable mobile robots to accurately locate tagged objects without requiring bulky infrastructure — work that has collectively attracted over 24 citations. His research demonstrates a strong commitment to building lightweight, deployable solutions that connect perception, localization, and real-world robotic applications.

Research Focus

Key Achievements

4
H-Index
6
Papers
72
Total Citations
12
Avg Citations/Paper
🏆 Most Cited Paper
MonoATT: Online Monocular 3D Object Detection with Adaptive Token Transformer
37 citations · 2023
📈 Most Prolific Year: 2023 (3 Papers)
🤝 Key Collaborators: 14
🏛 Institutions: Shanghai Jiao Tong University

Top Papers

  1. 1
  2. 2
  3. 3
  4. 4
  5. 5
  6. 6

Key Collaborators

Contact & Links

Available for collaboration
Content generated · 14 days ago