Recognition of Human Actions Based on Temporal Motion Templates
Samy Bakheet *
Department of Mathematics and Computer Science, Faculty of Science, Sohag University, P.O.Box 82524 Sohag, Egypt and Institute for Information Technology and Communications, Otto-von-Guericke-University Magdeburg, P.O.Box 4120, 39016 Magdeburg, Germany
Ayoub Al-Hamadi
Institute for Information Technology and Communications, Otto-von-Guericke-University Magdeburg, P.O.Box 4120, 39016 Magdeburg, Germany
M. A. Mofaddel
Department of Mathematics and Computer Science, Faculty of Science, Sohag University, P.O.Box 82524 Sohag, Egypt
*Author to whom correspondence should be addressed.
Abstract
Despite their attractive properties of invariance, robustness and reliability, statistical motion descriptions from temporal templates have not apparently received the amount of attention they might deserve in the human action recognition literature. In this paper, we propose an innovative approach for action recognition, where a novel fuzzy representation based on temporal motion templates is developed to model human actions as time series of low-dimensional descriptors. An NB (Naïve Bayes) classifier is trained on these features for action classification. When tested on a realistic action dataset incorporating a large collection of video data, the results demonstrate that the approach is able to achieve a recognition rate of as high as 93.7%, while remaining tractable for real-time operation.
Keywords: Human action recognition, temporal motion templates, naïve Bayes, IIKT action dataset, video interpretation