A Survey of Background Modeling Based on Robust Subspace Learning via Sparse and Low-rank Matrix Decomposition
Chaochao Xie, Yan Yan, Hanzi Wang, Lili Lin, Rui Wang · 2016
Moving object detection is an active and classical research topic in computer vision and pattern recognition. Background modeling based on robust subspace learning via sparse and low-rank matrix decomposition is currently one of the most popular frameworks to detect the moving objects in a video sequence, where a great lot of methods have been developed over the past few years. In this paper, we survey the background modeling methods based on robust subspace learning via sparse and low-rank matrix decomposition. We classify these robust background modeling methods in terms of different subspace learning formulation frameworks, and then analyze their respective pros and cons. Finally, we give the summary and prospects for future research.