Moving Object Detection via Robust Low Rank Matrix Decomposition with IRLS Scheme

Moving object detection is a key step in video surveillance system. Recently, Robust Principal Components Analysis (RPCA) shows a nice framework to separate moving objects from the background when the camera is fixed. The background sequence is then modeled by a low rank subspace that can gradually change over time, while the moving objects constitute the correlated sparse outliers. In this paper, we propose to use a low-rank matrix factorization with IRLS (Iteratively Reweighted Least Squares) scheme for RPCA decomposition and to address in the minimization process the spatial connexity of the pixels. Experimental results on different datasets show the pertinence of the proposed method.