|
Dr. Shih-Fu Chang
Professor
Digital Video and Multimedia Lab
Columbia University
November 21 at 3:00pm
106 Woodward
Abstract:
We are seeing an explosive growth of digital video in many domains, including Internet, mobile, enterprise, and consumer media. Today there are billions of videos online and about 80% of Internet traffic is related to video. This successful proliferation of video was driven by advances in video compression, transmission, and streaming. However, similar tools for discovering and searching information within large video archives are seriously lacking. In this talk, I will present an overview of recent research in video search, addressing some of the fundamental questions – what are the appropriate ontologies for defining concepts in video; is it possible to automate the detection of visual concepts by computer vision techniques with reasonable accuracy; how to combine visual content with textual keywords in video search; how to leverage visual matching with page rank-like ordering and visual pattern mining on Internet. Community resources and benchmarking efforts in related areas will also be discussed.
Bio:
Shih-Fu Chang is a Professor of Electrical Engineering and Director of Digital Video and Multimedia Lab at Columbia University. His research interests include image/video search, video content analysis, image forensics, and pervasive media. His group has developed several frequently cited content based visual retrieval systems (such as VisualSEEk and VideoQ) and more than 10 video indexing technologies transferred to industry. He will be given the IEEE Kiyo Tomiyasu Award in 2009 for his outstanding contributions to automatic image classification and search. Other recognitions of his include Navy ONR Young Investigator Award, IBM Faculty Award, NSF CAREER Award, and several Best Paper Awards from IEEE, ACM, and SPIE. He is an IEEE Fellow and Editor-in-Chief for IEEE Signal Processing Magazine (2006-8).
Back
|
|
|
Copyright © 2003
- 2008 College of Computing and Informatics
|
|
