Recognizing and Classifying Objects in Video Sequences: How close are we to being able to search for videos containing grandma?

Research Scientist
Multimedia Document Analysis Group
Ricoh California Research Center

Friday, February 15th, 3:00 PM, ENS 637

berna_erol@rii.ricoh.com

Abstract

Advancements in video object segmentation technology and the availability of efficient object-based video representations, such as MPEG-4, have resulted in the increased availability of arbitrarily shaped digital video content. While this enables many exciting applications, the process of locating and accessing a desired video sequence can still be challenging. In this talk, we will present the current technologies that are available for obtaining and representing object-based video, and then talk about techniques to recognize and classify these video objects using their shape, motion, and color information.

More information about the speaker and her relevant publications can be found at:

http://www.rii.ricoh.com/~berna/

For more information about Ricoh California Research Center, please visit:

http://www.rii.ricoh.com/ca_research_ctr.html

Biography

Berna Erol received her BSc degree in Control and Computer Engineering at Istanbul Technical University and her MASc and PhD degrees in Electrical and Computer Engineering at the University of British Columbia, Canada. Since September 2001, she has been a member of the Multimedia Document Analysis Group at Ricoh California Research Center. Her research interests include image and video compression, object-based video representations, and content-based visual information retrieval.

A list of Telecommunications and Signal Processing Seminars is available at from the ECE department Web pages under "Seminars". The Web address for the Telecommunications and Signal Processing Seminars is http://signal.ece.utexas.edu/seminars