Title: Beyond Deep Recognition: Discovering Visual Patterns in Big Visual Data

Speaker: Prof. Junsong Yuan (Associate Professor and Director of Visual Computing Lab of CSE Department, State University of New York at Buffalo)

Abstract: Thanks to the recent success of deep learning, many computer vision tasks nowadays are formulated as regression problems, from high-level vision tasks such as object detection and image captioning, to middle-level tasks such as structure from motion, and low-level tasks such as image registration and optical flow. However, often times one has to rely on large amounts of annotated training data to make the high-dimensional curve-fitting successful. In this talk, we will discuss a complementary yet overlooked problem beyond deep visual recognition and regression: discovering visual patterns in images and videos, and reveal that computer vision can be more than a regression problem. We will discuss recent work of unsupervised and weakly-supervised visual pattern discovery, and explore how to utilize them to better summarize, search, and interpret big visual data. Applications in object instance search, object detection, action recognition, and video summarization will also be discussed.

Bio: Junsong Yuan is an Associate Professor and Director of Visual Computing Lab of CSE Department, State University of New York at Buffalo. Before that he was an Associate Professor at Nanyang Technological University (NTU), Singapore, and he received his PhD from Northwestern University in 2009. He is currently Associate Editor of IEEE Trans. on Image Processing (T-IP) and IEEE Trans. on Circuits and Systems for Video Technology (T-CSVT), and served as area chair for CVPR/ACM MM/WACV/ACCV etc. He received 2016 Best Paper Award from IEEE Trans. on Multimedia, Nanyang Assistant Professorship from NTU, and Outstanding EECS Ph.D. Thesis award from Northwestern University. He is a Fellow of International Association of Pattern Recognition (IAPR).