News Story
Chellappa Gives Invited Talk at UC Berkeley

Prof. Rama Chellappa
In his talk, Prof. Chellappa discussed general principles for designing robust video-based pattern recognition systems, citing examples of statistical techniques for video-based modeling and recognition of human actions using motion, shape and behavior as features. He also spoke about methods for addressing variations due to time warping and viewpoints for action recognition and unsupervised clustering of video sequences.
More information can be found on the UC Berkeley website.
Published April 25, 2010