Hengshuang Zhao

Postdoctoral Researcher

Department of Engineering Science
University of Oxford
Parks Road, Oxford, OX1 3PJ
Email: hengshuangzhao [at] gmail.com or hengshuang.zhao [at] eng.ox.ac.uk


Biography

I am currently a Postdoctoral Researcher at the Torr Vision Group in the Department of Engineering Science at the University of Oxford, working with Prof. Philip Torr. Before that, I obtained my Ph.D. degree in the Department of Computer Science and Engineering at The Chinese University of Hong Kong , supervised by Prof. Jiaya Jia. During Ph.D., I have spent wonderful times as an intern with Dr. Jianping Shi, Prof. Xiaogang Wang at SenseTime (Beijing), Dr. Xiaohui Shen, Dr. Zhe Lin, Dr. Kalyan Sunkavalli, Dr. Brian Price at Adobe (San Jose), Prof. Raquel Urtasun at Uber (Toronto), and Dr. Vladlen Koltun at Intel (Santa Clara).

My general research interests cover the broad area of computer vision, machine learning and artificial intelligence, with special emphasis on building intelligent visual systems. My research goal is to utilize artificial intelligence techniques to make machines perceive, understand and interact with the surrounding environment, and ultimately make high positive impacts on various fields.

Update: I will be joining the Department of Computer Science at The University of Hong Kong as an Assistant Professor. I am looking for self-motivated Ph.D./RA/Interns to work together on computer vision and machine learning. Besides, several Postdoc positions with highly competitive salaries are waiting for talented candidates. In the meantime, our group at the University of Oxford is also looking for self-motivated interns and visitors to work on research projects related to 2D/3D scene recognition and reconstruction. If you are interested in joining my group at HKU or Oxford, please do not hesitate to drop me an email with your resume. Remote collaboration is also welcome.

Pinned: Highly optimized PyTorch codebases available for semantic segmentation semseg (PSPNet&PSANet).

Unified panoptic segmentation UPSNet.

Fully self-attention based image recognition SAN.

Powerful few-shot segmentation PFENet.

Simple, strong and efficient panoptic segmentation PanopticFCN.

Bottom up 3D instance segmentation PointGroup.

Backbone structure for 3D scene recognition Point Transformer.

Joint Modeling for 2D-3D scene recognition BPNet.

Publications [Google Scholar]

Experiences

Professional Activities

Talks & Presentations

Honors & Awards

Patents

Teaching

© Hengshuang Zhao | Last updated: 09/06/2021