Topic：Towards Deep Visual Scene Understanding
Time：2017年1月10日（周二）下午2:00 - 3:30
Venue：玉泉校区教一103会议室
Speaker：Prof. Michael Ying Yang,University of Twente, Netherlands
Bio:Michael Ying Yang is currently Assistant Professor with University of Twente (the Netherlands), heading a group working on scene understanding. He received the PhD degree (summa cum laude) from University of Bonn (Germany) in 2011. From 2008 to 2012, he worked as Researcher with the Department of Photogrammetry, University of Bonn. From 2012 to 2015, he was a Postdoctoral Researcher with the Institute for Information Processing, Leibniz University Hannover. From 2015 to 2016, he was a Senior Researcher with Computer Vision Lab Dresden, TU Dresden. His research interests are in the fields of computer vision and photogrammetry with specialization on scene understanding and semantic interpretation from imagery and videos. He published over 50 articles in international journals and conference proceedings and currently co-supervise 3 PhD students. He serves as Associate Editor of ASPRS Photogrammetric Engineering & Remote Sensing, co-chair of ISPRS working group II/5 Dynamic Scene Analysis, and recipient of the ISPRS President's Honorary Citation (2016) and Best Science Paper Award at BMVC 2016. He co-organized 6 workshops with CVPR/ICCV/ECCV, and is guest editor of 4 journal special issues. Since 2016, he is a Senior Member of IEEE. He is regularly serving as program committee member of conferences and reviewer for international journals.
Abstract:
Inspired by the ability of humans to interpret and understand visual scenes nearly effortlessly, the problem of visual scene understanding has long been advocated as the "holy grail" of computer vision. In recent years there has been considerable progress on many sub-problems of the overall scene understanding problem. Due to the rise of deep learning, the performance for these sub-tasks starts to achieve remarkable performance levels. This talk highlights recent progress on some essential components such as semantic segmentation and pose estimation, as well as on our work on semantic scene graph. These efforts are part of a longer-term agenda towards deep visual scene understanding.