Hengshuang Zhao

Dr. Hengshuang Zhao is an Assistant Professor at the Department of Computer Science, University of Hong Kong. He holds a PhD from The Chinese University of Hong Kong and completed postdoctoral research at MIT CSAIL and the University of Oxford. His research focuses on computer vision, machine learning, and AI, emphasizing visual scene understanding, generative modeling, autonomous driving, and embodied AI. He leads a lab actively recruiting PhD students and researchers.

• Education:
• PhD in Computer Science, Chinese University of Hong Kong (2017)
• Bachelor's Degree, Huazhong University of Science and Technology
• Affiliations:
• Member of Hong Kong Robotics Institute
• Organizer of CVPR/ICML workshops on point clouds and embodied AI

His research interests include: Scene Understanding (e.g., semantic segmentation, 3D reconstruction), Multimodal Learning (vision-language models), and Generative Modeling (image/video synthesis). Recent work includes Depth Anything (CVPR 2024 Best Demo), Point Transformer V3 (CVPR 2024), and DriveGPT4 series for autonomous driving.

He has been recognized with awards including the Excellent Young Scientists Fund (2024), AI 2000 Influential Scholar in CV (2022-2024), and organized multiple top-tier conferences' workshops and tutorials.