Volume 9, Issue 4, December 2014, Pages 1708–1717
Haiam A. Abdul-Azim1, Elsayed E. Hemayed2, and Magda B. Fayek3
1 Physics Department, Faculty of Women for Arts, Science and Education, Ain Shams University, Cairo, Egypt
2 Computer Engineering Department, Faculty of Engineering, Cairo University, Giza, Egypt
3 Computer Engineering Department, Faculty of Engineering, Cairo University, Giza, Egypt
Original language: English
Copyright © 2014 ISSR Journals. This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.
Human action recognition remains a challenging problem for researchers. Several action representation approaches have been proposed to improve the action recognition performance. Recently, local space-time features have become a popular representation approach for human actions in video sequences. Many different space-time detectors and descriptors have been proposed. They are evaluated on different datasets using different experimental conditions. In this paper, the performance of Cuboid detector is evaluated with four space-time description methods; namely, Gradient, HOG, HOF and HOG-HOF. All descriptors were tested on two datasets (KTH and Weizmann) using the bag-of-words model and Support Vector Machine.
Author Keywords: Space-time features, Cuboid detector, space-time feature descriptors, bag-of-words, human action recognition.
Haiam A. Abdul-Azim1, Elsayed E. Hemayed2, and Magda B. Fayek3
1 Physics Department, Faculty of Women for Arts, Science and Education, Ain Shams University, Cairo, Egypt
2 Computer Engineering Department, Faculty of Engineering, Cairo University, Giza, Egypt
3 Computer Engineering Department, Faculty of Engineering, Cairo University, Giza, Egypt
Original language: English
Copyright © 2014 ISSR Journals. This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.
Abstract
Human action recognition remains a challenging problem for researchers. Several action representation approaches have been proposed to improve the action recognition performance. Recently, local space-time features have become a popular representation approach for human actions in video sequences. Many different space-time detectors and descriptors have been proposed. They are evaluated on different datasets using different experimental conditions. In this paper, the performance of Cuboid detector is evaluated with four space-time description methods; namely, Gradient, HOG, HOF and HOG-HOF. All descriptors were tested on two datasets (KTH and Weizmann) using the bag-of-words model and Support Vector Machine.
Author Keywords: Space-time features, Cuboid detector, space-time feature descriptors, bag-of-words, human action recognition.
How to Cite this Article
Haiam A. Abdul-Azim, Elsayed E. Hemayed, and Magda B. Fayek, “Evaluation of Local Space-time Descriptors based on Cuboid Detector in Human Action Recognition,” International Journal of Innovation and Applied Studies, vol. 9, no. 4, pp. 1708–1717, December 2014.
