Jin PAN

jpan AT link DOT cuhk DOT edu DOT hk

About

I am a senior researcher led by Tong Qin in Intelligent Automative Solution(IAS) BU, Huawei. In 2022, I received my Ph.D. degree in Robotics, Perception & AI Lab (Rπ Lab) in The Electronic Engineering Department at The Chinese University of Hong Kong, supervised by Professor Max Q.-H. Meng. Before CUHK, I received a B.Eng. degree from Xi'an Jiaotong University

My research interests lie in Autonomous Driving, Robotics and SLAM. Now I focus on how to robustly establish the correspondences from different modalities. 



News

Publications

A Large-Scale Dataset for Benchmarking Elevator Button Segmentation and Character Recognition

Jianbang Liu, Yuqi Fang, Delong Zhu, Nachuan Ma, Jin Pan, Max Q.-H. Meng

(IEEE International Conference on Robotics and Automation, ICRA 2021)

Globally Learnable Point Set Registration Between 3D CT and Multi-View 2D X-Ray Images of Hip Phantom 

Jin Pan, Zhe Min, Ang Zhang, Han Ma, and Max Q.-H. Meng

(2021 IEEE International Conference on Robotics and Biomimetics, ROBIO 2021)

A Nonuniform Sampling Strategy for Path Planning Using Heuristic-based Certificate Set. 

Han Ma, Jianbang Liu, Fei Meng, Jin Pan, Jiankun Wang, Max Q.-H. Meng. 

(2021 IEEE International Conference on Robotics and Biomimetics, ROBIO 2021)

Framework for Airport Trolley Deployment Robot

A Searching Space Constrained Partial to Full Registration Approach with Applications in Airport Trolley Deployment Robot

Jin Pan, Xiaochun Mai, Chaoqun Wang, Zhe Min, Jiankun Wang, Hu Cheng, Tingguang Li, Erli Lyu, Li Liu and Max Q.-H. Meng

(IEEE Sensors Journal 2020)

Robust and Accurate Point Set Registration with Generalized Bayesian Coherent Point Drift

Ang Zhang, Zhe Min, Jin Pan, Max Q.-H. Meng

(IEEE/RSJ International Conference on Intelligent Robots and Systems, IROS 2021)

Generalized 3D Point Set Registration with Hybrid Mixture Models for Computer-Assisted Orthopaedic Surgery

Zhe Min, Jiaole Wang, Jin Pan, Max Q.-H. Meng

(IEEE Transactions on Automation Science and Engineering, IEEE TASE 2020)

Paper

Coarse-to-Fine Visual Object Catching Strategy Applied in Autonomous Airport Baggage Trolley Collection

Chaoqun Wang, Xiaochun Mai, Danny Ho, Tingting Liu, Chenming Li, Jin Pan, Max Q.-H. Meng

(IEEE Sensors Journal 2020)

Paper

Double-view 2D-3D Registration on Cerebral Angiograms 

Double-View Global 2D-3D Registration Based on Branch and Bound Algorithm

Jin Pan, Zhe Min and Max Q.-H. Meng (IEEE International Symposium on Biomedical Imaging, one-page paper, ISBI 2020)

Multi-view 2D-3D Registration 

Multi-View Global 2D-3D Registration Based on Branch and Bound Algorithm

Jin Pan, Zhe Min, Ang Zhang, Han Ma and Max Q.-H. Meng (IEEE International Conference on Robotics and Biomimetics, ROBIO 2019)

Paper

Non-Rigid Registration with Anisotropic Uncertainties

Robust Non-Rigid Point Set Registration Algorithm Considering Anisotropic Uncertainties Based on Coherent Point Drift

Zhe Min, Jin Pan, Ang Zhang, Max Q.-H. Meng (IEEE/RSJ International Conference on Intelligent Robots and Systems, IROS 2019)

Paper

Hierarchical Deep Reinforcement Learning Framework

Learning to Interrupt: A Hierarchical Deep Reinforcement Learning Framework for Efficient Exploration

Tingguang Li, Jin Pan, Delong Zhu, Max Q.-H. Meng.

(IEEE International Conference on Robotics and Biomimetics, ROBIO 2018)

Paper

Honors

Challenges

ICRA 2017 DJI RoboMaster Manipulation Challenge in Singapore (Finalist, 5/93)

Organized by DJI and IEEE ICRA, The ground robot challenge examines the application and competence of technologies that include positioning, object grasping, force control, target identification, and system stability. Teams are challenged to develop a lightweight mobile manipulator that can autonomously pick, transport and stack building blocks. 

We developed a mobile manipulator which can autonomously detect, pick, transport, and stack the target cubes. In the final competition, we successfully piled up 10 cubes and ranked 5th place from 93 teams all over the world.

Services



Teaching