Summary: In Proceedings of the Third International Conference on Automatic Face and Gesture Recognition, pp. 10­15, Nara, Japan
Extracting Gestural Motion Trajectories
Ming­Hsuan Yang and Narendra Ahuja
Beckman Institute and Department of Electrical and Computer Engineering
University of Illinois at Urbana­Champaign
Urbana, IL 61801
fmhyang, ahujag@stereo.ai.uiuc.edu
Abstract
This paper is concerned with the extraction of spatio­
temporal patterns in video sequences with focus on trajec­
tories of gestural motions associated with American Sign
Language. An algorithm is described to extract the motion
trajectories of salient features such as human palms from
an image sequence. First, motion segmentation of the im­
age sequence is generated based on a multiscale segmen­
tation of the frames and attributed graph matching of re­
gions across frames. This produces region correspondences
and their affine transformations. Second, colors of the mov­
ing regions are used to determine skin regions. Third, the
head and palm regions are identified based on the shape and

Source: Ahuja, Narendra - Department of Electrical and Computer Engineering, University of Illinois at Urbana-Champaign

Collections: Computer Technologies and Information Sciences; Engineering