Sanjeev R. Kulkarni

Professor of Electrical Engineering
Ph.D. 1991, Massachusetts Institute of Technology

My research is focused primarily in two main areas: adaptive and learning systems, and image/video processing. In addition to these areas (described below), my work has also contributed to the areas of information theory and statistics.

The use of adaptive and learning strategies in engineering systems is important for several reasons, including increased robustness against changes in operating conditions and errors in data and modeling, reduced need for explicit programming, and improved performance over time. Although such strat egies have been widely used and studied, certain basic questions such as "What are optimal strategies for adaptation?" and "How much data is required?" remain largely unanswered. Therefore, the design of adaptive and learning systems i s generally done in an ad hoc manner with little experience gained from one application to the next. Furthermore, very few performance guarantees can be provided for the algorithms used.

The goal of my research in this area is to contribute to the theoretical foundations of adaptive and learning systems. Particular topics under investigation include learning theory, statistical pattern recognition, nonparametric estimation, and system identification. The primary goal is to obtain results on the fundamental capabilities and limitations of adaptive and learning systems in order to allow more principled design and analysis of such systems.

I have also been involved in research in several areas of image and video processing. This research includes work on the geometric problems in image processing and video analysis. The work on geometric problems includes curve/boundary representation, convex set reconstruction, and the use of results from learning theory to provide sample size bounds on the amount of data needed for various problems in geometric reconstruction. My work on video analysis includes techniques for creating panoramic views from a set of images, and annotation and browsing of video from compressed data.