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.

‚Äč