Action classification algorithm based on EGEI and LPP

A foreground extraction algorithm based on background subtraction and edge detection was proposed to obtain the foreground with a little change. An action classification method based on Enhanced Gait Energy Image (EGEI) and Locality Preserving Projections (LPP) was used. The high dimensional feature space was non-linearly reduced to lower dimensional space, which outperformed PCA and 2DPCA. The nearest-neighbor classifier was adopted to distinguish different actions. This algorithm needn´t extract the period of the video, which was indispensable in some other methods. Experimental results show that the algorithm is simple, and achieves higher classification accuracy with less running time.