Seminar by Dr. Subrata Rakshit
Content Based Image Retrieval System
Dr. Subrata Rakshit
Centre for Artificial Intelligence & Robotics
Bangalore
Date: Thursday, February 06, 2003
Time: 03:45 PM
Venue: CS-101
Abstract
Content based image retrieval has become a topic of interest as digital image archive build up. In particular, such systems will be of interest to defense users as they try to cope with an increasing flood of imagery based intelligence. CAIR has been trying to build CBIR systems for such users leveraging the techniques already available in the literature. The talk will highlight the issues that we tried to address, the way we have combined available techniques and the improvisations that we have done. The present bottlenecks and avenues for further work will also be addressed. Finally, a demonstration of the current software will be given to illustrate the types of queries that can be handled at present.
About the Speaker
Dr. Subrata Rakshit received his B.Tech. in Engineering Physics from IIT Bombay in 1988. He completed his MS degree in Electrical Engineering in 1989, and PhD in the area of Image Processing in 1994, both at California Institute of Technology, Pasadena. He briefly worked as a post-doctoral fellow in the Department of Anatomy and Neurobiology of Washington University Medical School at St. Louis. He joined the Centre for Artificial Intelligence and Robotics (CAIR), a DRDO laboratory, in Dec 1994, where he heads the Computer Vision Group. His research interests are in image processing, multiresolution, information coding in images and multisensor data fusion. Recent projects have been in the areas of satellite based change detection, content based image retrieval and progressive image transmission for unreliable, low bandwidth channels.