I am currently a professor in the Computer Science and Engg. dept. at IIT Madras heading the Computer Vision Lab. Prior to this, from 2002-2005, I was a Research Scientist at the Real-Time Vision and Modeling Department at Siemens Corporate Research, Princeton NJ. I completed my PhD in Computer Science from the University of Maryland, College Park in Dec 2002. Before that, I completed an MS from Cornell University in 2000 and a B.Tech. in Computer Science and Engineering from the Indian Institute of Technology, Delhi in 1997.
My research interests are in all areas of Computer Vision, with special interest in combining geometric and statistical inference techniques to build robust Computer Vision systems. This often involves consideration of illumination- and other invariants to build systems robust to different variations in the data and their statistical analysis (credit due to V. Ramesh at Siemens here).
I also have significant interest in shape as an important cue for Image Understanding and several of my recent works have developed both theoretical and practical systems for shape-based detection, matching and retrieval. Most recently, we developed the shape-based CoMaL point detector, matcher and tracker that significantly improves upon the Harris/KLT at the object boundaries.
More recently, we have been working on Convolutional Neural Networks and in particular, their applications to various Computer Vision Problems. Some of our recent works include analysis and algorithms for zero/low-shot scenarios in different Computer Vision tasks such as image classification. These often involve an interplay between text and image data, and some of the latest techniques for image understanding and generation such as VAEs and GANs.