Spatial-Temporal Sparse Representation for Background Modeling

Abstract

In this paper, a sparse representation based background model is introduced for video surveillance. Inspired by the fact that spatial and temporal information are both important for foreground detection, a spatial-temporal image patch, namely brick, is used as atomic unit for online subspace learning and sparse representation. Furthermore, Random Projection emerged from Compressive Sensing theory is applied to reduce the dimension of bricks so as to speed up the algorithm. Experimental results show the effectiveness of the proposed method.