August 9, 2004

BACK   

Recognition of humans using face and gait is gaining interest in security and passport control applications. I will present some new results on face recognition from still and video images. The still image algorithm relies on a generalized stereo approach and can handle illuminations variations and constrained pose variations. The video-based algorithm computes the probability of finding a face in a given video using particle filters. We will also discuss the state-of the art in recognizing humans at a distance using gait. I will conclude the talk by discussing two methods for modeling motion trajectories using factorization theorem and statistical shape models with applications to human activity modeling and anomaly detection.

Rama Chellappa is a professor of Electrical and Computer engineering and an affiliate professor in the Department of Computer Science at the University of Maryland. He is also a permanent member of the University of Maryland Institute for Advanced Computer studies and the Director of Center for Automation Research. Over the last twenty three years, he has contributed to recovering 3D models of environment using still visible and radar images, recovering 3D shape using structure from motion techniques, recognizing humans using face and gait information, nand modeling and recognizing the activities of one or more humans in the scene. In addition to developing algorithms for accomplishing the tasks mentioned above, he has also developed theoretical underpinnings for existence, uniqueness of solutions and performance bounds. For his work in this field, Dr. Chellappa was elected as a fellow of the Institute of Electrical and Electronics Engineers (IEEE) and the International Association of Pattern Recognition. In addition, he has received many research and teaching awards. He received the 1985 NSF Presidential Young Investigator award, one of the 1985 IBM faculty development awards, Excellence in Teaching award from the School of Engineering, USC, in 1990, the Best Conference Paper from the International Association of Pattern Recognition in 1992, and the IEEE Signal Processing Society Technical Achievement Award in 2001. He was elected as a Distinguished Research Faculty Fellow in 1996 and as one of five Distinguished Scholar-Teachers in 2003. He has served on the editorial boards of IEEE Transactions on Acoustics, Speech and Signal Processing,Image Processing, Neural Networks and Pattern Analysis and Machine Intelligence. Currently he is serving as the Editor-in-Chief of the IEEE Transactions on Pattern Analysis and Machine Intelligence. He is also serving as the Vice-President of IEEE Signal processing Society for Awards.

BACK

Last Updated: Aug. 9, 2004