FOREGROUND DETECTION BASED ON LOW-RANK AND BLOCK-SPARSE MATRIX DECOMPOSITION

Foreground detection is the first step in video surveillance system to detect moving objects. Principal Components Analysis (PCA) shows a nice framework to separate moving objects from the background but without a mechanism of robust analysis, the moving objects may be absorbed into the background model. This drawback can be solved by recent researches on Robust Principal Component Analysis (RPCA). The background sequence is then modeled by a low rank sub-space that can gradually change over time, while the moving foreground objects constitute the correlated sparse outliers. In this paper, we propose to use a RPCA method based on low-rank and block-sparse matrix decomposition to achieve foreground detection. This decomposition enforces the low-rankness of the background and the block-sparsity aspect of the foreground. Experimental results on different datasets show the pertinence of the proposed approach.