Background/Foreground Separation for Static RGB Cameras

Applying RPCA via decomposition in low rank and sparse matrices in background/foreground separation, the background sequence is modeled by the low-rank subspace that can gradually change over time, while the moving foreground objects constitute the correlated sparse outliers. For example, Figure 1 shows original frames of sequences from the BMC dataset and its decomposition into the low-rank matrix L and sparse matrix S. We can see that L corresponds to the background whereas S corresponds to the foreground. The fourth image shows the foreground mask obtained by thresholding the matrix S. So, the different advances in the different frameworks of the decomposition in low rank and additive matrices are fundamental and can be applied to background modeling and foreground detection in video surveillance. (See also Background Subtraction Website).

Figure 1: RPCA via decomposition in low rank and sparse matrices in background/foreground

separation: Original image, low-rank matrix L (background), sparse matrix S (foreground),

foreground mask (Sequences from BMC 2012 dataset.

Publications

T. Bouwmans . A. Sobral, S. Javed, S. Jung, E. Zahzah, "Decomposition into Low-rank plus Additive Matrices for Background/Foreground Separation: A Review for a Comparative Evaluation with a Large-Scale Dataset", Computer Science Review, November 2016.

T. Bouwmans, E. Zahzah, “Robust PCA via Principal Component Pursuit: A Review for a Comparative Evaluation in Video Surveillance”, Special Issue on Background Models Challenge, Computer Vision and Image Understanding, CVIU 2014, Volume 122, pages 22–34, May 2014.

C. Guyon, T. Bouwmans, E. Zahzah, “Foreground Detection via Robust Low Rank Matrix Decomposition including Spatio-Temporal Constraint”, International Workshop on Background Model Challenges, ACCV 2012, pages 315-320, Daejeon, Korea, November 2012

C. Guyon, T. Bouwmans, E. Zahzah, "Foreground Detection by Robust PCA solved via a Linearized Alternating Direction Method", International Conference on Image Analysis and Recognition, ICIAR 2012, pages 115-122, Aveiro, Portugal, June 2012.

C. Guyon, T. Bouwmans, E. Zahzah, "Foreground detection based on low-rank and block-sparse matrix decomposition", IEEE International Conference on Image Processing, ICIP 2012, pages 1225-1228, Orlando, Florida, USA, September 2012.