Learning to Understand Visual Data with Minimal Human Supervision - Yong Jae Lee (University of California)

Host: Leonid Sigal

Abstract:

Humans and animals learn to see the world mostly on their own, without supervision, yet today’s state-of-the-art visual recognition systems rely on millions of manually-annotated training images. This reliance on labelled data has become one of the key bottlenecks in creating systems that can attain a human-level understanding of the vast concepts and complexities of our visual world. Indeed, while computer vision research has made tremendous progress, most success stories are limited to specific domains in which lots of carefully-labelled data can be unambiguously and easily acquired.

In this talk, I will present my research in computer vision and deep learning on creating scalable recognition systems that can learn to understand visual data with minimal human supervision. Given the right constraints, I’ll show that one can design learning algorithms that discover and generate meaningful patterns from the data with little to no human supervision. In particular, I’ll focus on algorithms that can localize relevant image regions given only weak image-level supervision, and hierarchically disentangle and generate fine-grained details of objects. I'll also discuss our latest work on real-time instance segmentation -- our algorithm obtains ~30 mAP on the challenging MS COCO dataset while running at 33 fps on a single Titan Xp.

Bio:

Yong Jae Lee is an Assistant Professor in the Department of Computer Science at the University of California, Davis. His research interests are in computer vision, machine learning, and computer graphics, with a focus on creating robust visual recognition systems that can learn to understand the visual world with minimal human supervision. Before joining UC Davis in 2014, he received his Ph.D. from the University of Texas at Austin in 2012 advised by Kristen Grauman, and was a post-doc at Carnegie Mellon University (2012-2013) and UC Berkeley (2013-2014) advised by Alyosha Efros. He received his B.S. in Electrical Engineering from the University of Illinois at Urbana-Champaign in 2006. He is a recipient of several awards including the Army Research Office (ARO) Young Investigator Program (YIP) award, UC Davis Hellman Foundation Fellowship, National Science Foundation (NSF) CAREER award, AWS Machine Learning Research Award, Adobe Data Science Research Award, and UC Davis College of Engineering Outstanding Junior Faculty Award.