Zhenjun Zhao (赵祯俊) I received my PhD from the Chinese University of Hong Kong (CUHK), under the supervision of Ben M. Chen. I am currently working at Unmanned Systems Research Group lead by Ben M. Chen at CUHK. I have been fortunate to collaborate with others, including Haoang Li from HKUST (Guangzhou) , Chen Wang from University at Buffalo (UB), and Peidong Liu from Westlake University. Email  |  GitHub  |  Google Scholar  |  LinkedIn  |  Twitter

My research work mainly lies at the intersection of robotics and 3D computer vision, with the applications to robotics, autonomous driving, virtual reality and augmented reality etc. Currently I am exploring how to best integrate machine learning techniques (i.e. mainly deep learning) with classical geometry-based 3D visual perception pipelines to improve their performance, especially under extremely challenging conditions. In particular, design algorithms that better blend geometric inductive bias and powerful data-driven approaches.

• 07/2024 Our paper GlobalPointer is accepted to ECCV 2024!
• 04/2024 Our paper DeDoDe v2 is accepted to CVPR 2024 Workshop!
• 10/2023 Our paper BALF is accepted to WACV 2024!
• 04/2023 Our paper Hong Kong World is accepted to TPAMI!
• 04/2023 Our paper Benchmark for Evaluating Initialization of Visual-Inertial Odometry is accepted to CCC 2023!
• 01/2023 Our paper SyreaNet is accepted to ICRA 2023!

	GlobalPointer: Large-Scale Plane Adjustment with Bi-Convex Relaxation Bangyan Liao*, Zhenjun Zhao*, Lu Chen, Haoang Li, Daniel Cremers, Peidong Liu European Conference on Computer Vision (ECCV), 2024 paper | project page | code A globally optimal and efficient large-scale plane adjustment, using alternating minimization and convex relaxation techniques.
	DeDoDe v2: Analyzing and Improving the DeDoDe Keypoint Detector Johan Edstedt, Georg Bökman, Zhenjun Zhao IEEE/CVF Computer Vision and Pattern Recognition Conference (CVPR) Workshop on Image Matching: Local Features & Beyond, 2024 paper | slides | code An improved keypoint detector built on top of DeDoDe.
	BALF: Simple and Efficient Blur Aware Local Feature Detector Zhenjun Zhao IEEE/CVF Winter Conference on Applications of Computer Vision (WACV), 2024 paper | project page | video | code A simple yet both efficient and effective motion blur aware local feature detector.
	Benchmark for Evaluating Initialization of Visual-Inertial Odometry Zhenjun Zhao, Ben M. Chen Chinese Control Conference (CCC), 2023 paper | code A novel benchmark for the evaluation of the initialization of visual-inertial odometry (VIO).
	Hong Kong World: Leveraging Structural Regularity for Line-based SLAM Haoang Li, Ji Zhao, Jean-Charles Bazin, Pyojin Kim, Kyungdon Joo, Zhenjun Zhao, Yun-Hui Liu IEEE Transactions on Pattern Analysis and Machine Intelligence (TPAMI), 2023 paper | code (coming soon) A novel structural model called Hong Kong world to describe the structured scenes with vertical, horizontal and sloping dominant directions.
	SyreaNet: A Physically Guided Underwater Image Enhancement Framework Integrating Synthetic and Real Images Junjie Wen, Jinqiang Cui, Zhenjun Zhao, Ruixin Yan, Zhi Gao, Lihua Dou, Ben M. Chen IEEE International Conference on Robotics and Automation (ICRA), 2023 paper | poster | code A novel UIE framework combining both synthetic and real data under the guidance of the revised underwater image formation model and DA strategies.

Co-supervisor, Undergraduate Final Year Project, 2019-2020 Co-supervisor, MSc Project, 2019-2020 Teaching Assistant, Multivariable Calculus for Engineers, Spring 2021 Teaching Assistant, Complex Variables for Engineers, Fall 2020 Teaching Assistant, Probability and Statistics for Engineers, Spring 2020 Teaching Assistant (Lead), Introduction to Control Systems, Fall 2019