Detecting moving objects from dynamic background combining subspace learning with mixed norm approach

Abstract

Detecting moving objects from the video is an important research topic in computer vision. In this paper, we first review some traditional methods for moving objects detection. Then, some related preliminaries about the subspace learning and the mixed norm are introduced due to their remarkable advantages shown in signal processing. Combing the mixed norm and the total variation, an effective regularized framework is proposed to achieve more pure foreground. Furthermore, by introducing the subspace learning, the model has lower computational complexity. An efficient method based on the alternating direction method of multipliers (ADMM) is then developed to deal with the proposed constrained minimization problem. Extensive experiments on the private and real-world datasets show that the proposed approach outperforms the existing state-of-the-art approaches, particularly for the cases with dynamic background, which indicates the worth to integrate the subspace learning, the l2,1 mixed norm and total variation into a low-rank representation model.