Title :
Simultaneous Human detection and action recognition employing 2DPCA-HOG
Author :
Naiel, Mohamed A. ; Abdelwahab, Moataz M. ; Elsaban, M. ; Mikhael, Wasfy B.
Author_Institution :
Nile Univ., Sixth October, Egypt
Abstract :
In this paper a novel algorithm for Human detection and action recognition in videos is presented. The algorithm is based on Two-Dimensional Principal Components Analysis (2DPCA) applied to Histogram of Oriented Gradients (HOG). Due to simultaneous Human detection and action recognition employing the same algorithm, the computational complexity is reduced to a great deal. Experimental results applied to public datasets confirm these excellent properties compared to most recent methods.
Keywords :
computational complexity; object recognition; principal component analysis; video surveillance; 2DPCA-HOG; action recognition; computational complexity; histogram of oriented gradients; human detection; public datasets; two-dimensional principal components analysis; videos; Accuracy; Cameras; Humans;
Conference_Titel :
Circuits and Systems (MWSCAS), 2011 IEEE 54th International Midwest Symposium on
Conference_Location :
Seoul
Print_ISBN :
978-1-61284-856-3
Electronic_ISBN :
1548-3746
DOI :
10.1109/MWSCAS.2011.6026441
