Published more than 400 articles in top-tier conferences and journals with over 60,000 citations. Received the CVPR 2023 Best Paper Award and AAAI 2021 Outstanding Paper Award. Developed several important open-source projects including InternImage (the strongest open-source visual universal backbone model) and InternVideo (one of the earliest and strongest open-sourced video understanding backbone models).
Research Experience
Professor Yu Qiao and his team invented center loss and temporal segment networks. They have focused on this field for over 8 years, creating innovative methodologies and models such as Unmasked Teacher, Uniformer, and Videomae.
Education
Served as an assistant professor at the Graduate School of Information Science and Technology, University of Tokyo from 2009 to 2010. He has been working on deep learning since 2006 and is one of the earliest people to introduce deep learning to video understanding.
Background
Professor Yu Qiao is the lead scientist at Shanghai AI Laboratory and an honored director of the Multimedia Laboratory at Shenzhen Institutes of Advanced Technology, Chinese Academy of Sciences (SIAT). His research interests revolve around foundation models, computer vision, deep learning, robotics, and AI applications.