Biograph: Jinhui Tang received the B.Eng. and Ph.D. degrees from the University of Science and Technology of China in 2003 and 2008, respectively. He is currently a Professor at the Nanjing University of Science and Technology. He has authored over 150 papers in top-tier journals and conferences. His research interests include multimedia analysis and computer vision. He was a recipient of the best paper awards in ACM MM 2007, PCM 2011 and ICIMCS 2011, the Best Paper Runner-up in ACM MM 2015, and the best student paper awards in MMM 2016 and ICIMCS 2017. He has served as an Associate Editor for the IEEE TNNLS, IEEE TKDE, IEEE TMM and IEEE TCSVT.
Title: Weakly Supervised Image Understanding
Abstract: The increasing social media websites allow users not only to create and share media data but also to rate and annotate them. Thus lots of meta-data, such as user-provided tags, comments, geo-tags, capture time and EXIF information, associated to multimedia resources, are available in the WWW. However, these meta-data has a lot of noise. In this talk, I will introduce some recent research works on how to utilize the weakly supervised information to learn visual semantics, including weakly supervised transfer network for image understanding and tensor based social tag refinement.