Seminar by Dr. Visvanathan Ramesh
Performance Measures for Real-Time Vision Systems
Dr. Visvanathan Ramesh
Siemens Corporate Research
Princeton, NJ
USA
Date: Mon, Sep 27, 2004
Time: 3:00 PM
Venue: CS-101
Abstract
Focuses on statistical algorithms for Computer Vision, and how real-time performance in such situations depends on the entropy of the input data. He will also give an overview of Vision research in the 100-member team at Siemens Corporate Research.
About the Speaker
Dr. Visvanathan Ramesh obtained his doctoral degree from the Department of EE at the University of Washington, where he defended his Ph.D dissertation titled "Performance Characterization of Image Understanding algorithms" in 1994. He has been actively involved in Image and Video Understanding research in low and mid level vision over the past 12 years and has published numerous publications in the topic. His primary objective is to build robust image and video analysis systems and to quantify robustness of IU algorithms. Dr. Ramesh is currently a senior member of technical staff and project manager of the real-time imaging effort in the Imaging department at Siemens Corporate Research in Princeton NJ. At Siemens, he has focused on the research and development of statistical method for real-time video analysis functions such as object detection, tracking, and action recognition. He is a co-author of a paper on real-time tracking that rece the best paper award in CVPR 2000. His broad research interests are Pattern Recognition, Computer Vision, AI and Biomedical Engineering.