"CHITRA" is a Sanskrit word that means 'picture' and I work in the area of analyzing pictures --- images and video. I am a Research Staff Member in the Internet Infrastructure and Computing Utilities Department at the IBM Thomas J. Watson Research Center, located in Hawthorne, New York where I lead the E-learning Media and MediaSemantics Projects. I also serve as the Research Relationship Manager for Media.

My research interests and consequently specific projects over the years have spanned Multimedia Systems and Digital Video Analysis, Computer Vision and Image Processing, Pattern Recognition and Data Clustering, Machine Learning, and Computer Graphics.

During my Ph.D. program at the Department of Computer Science, Michigan State University, I studied new means to represent and recognize 3-D free-form objects using dense range data and proposed COSMOS, a powerful surface representation scheme to represent and recognize general, arbitrarily curved objects. I also worked on view registration algorithms for the estimation of pose of an object and for automatic model construction of complex shapes from multiple views. All archival publications related to these aspects of my research have appeared in the IEEE Transactions of Pattern Analysis and Machine Intelligence .

My current research focuses on developing technologies for automatic digital video analysis in various domains such as education and training media, motion pictures, and news, that are useful in content-based structuralization, annotation and search, and smart browsing. See Publications for my papers in this area.

First, as part of the VideoVista project, which was partially supported by the NIST Advanced Technology Program in Digital Video, I developed the first HDTV video content management system that automatically analyzes motion in MPEG-2 encoded videos within the compressed domain itself, and produces video descriptors characterizing the perceived visual motion for content-based retrieval applications in a digital television studio. The results of this system have been demonstrated at the NAB Exhibition in 1998. The compressed domain motion analyzer features in the IBM system that competed in NIST Video-TREC, 2001.

I also built a novel videotext extraction and recognition system for content-based annotation and retrieval of videos. This videotext work received the Best Industrial-related Award at the 1998 International Conference on Pattern Recognition. A multimedia description scheme based on video text, contributed by us, has recently become part of the MPEG-7 standard. In a related context, I also investigate extraction of events, significant happenings in video for efficient video ToC construction and nonlinear browsing.

Recently, I've created a new research approach, jointly with Svetha Venkatesh, called the Computational Media Aesthetics to address the problem of semantic gap in automatic content annotation systems. Computational media aesthetics is defined as the algorithmic study of a variety of image and aural elements in media, founded on their patterns of use in film grammar, and the computational analysis of the principles that have emerged underlying their manipulation, individually or jointly, in the creative art of clarifying, intensifying, and interpreting some event for the audience. See IEEE Multimedia, Oct-Dec 2001 issue for our article on this.

This media production-guided semantic analysis approach has been well received in the research community. It has been recognized with awards and selection such as: Top-ranked paper in ACCV 2002; Best Paper Prize in IEEE PRCM 2000; Edited volume titled Media Computing Computational Media Aesthetics published in June 2002 by Kluwer Academic Publishers; and Special Issue of IEEE Multimedia dedicated to this theme in Spring 2003. Click here for the Call for Papers for this special issue.

I enjoy collaborating with researchers all over the world, and some recent collaborators are Dr. Vikrant Kobla, University of Maryland, College Park, Maryland, USA, Prof. Jae-Chang Shim, Andong National University, Andong Kyungpook, South Korea, Prof. Svetha Venkatesh, Department of Computer Science, Curtin University of Technology, Perth, Western Australia, Dr. Kim Shearer, IDIAP, Martigny, Switzerland, Dr. Hrishikesh Aradhye, SRI, California, USA, Gaurav Jain, Department of Computer and Information Science, UPenn, Philadelphia, Pennsylvania, USA, and --- Brett Adams, Ba Tu Truong, Simon Moncrieff, Phungquo Dinh --- graduate students at the School of Computing in Curtin University of Technology, Perth, Western Australia.