An Efficient Matrix Factorization for Dynamic Background Subtraction

Abstract

Detection of moving objects in a video sequence is a difficult task and robust moving object detection in video frames for video surveillance applications is a challenging problem. Object detection is a fundamental step for automated video analysis in many vision applications. Object detection in a video is usually performed by object detectors or background subtraction techniques. We recommend an effective online background subtraction method, which can be robustly applied to practical videos that have variations in both foreground and background. Different from previous methods which often model the foreground as Gaussian or Laplacian distributions, we prototypical the foreground for each frame with a specific mixture of Gaussians (MoG) distribution, which is updated online frame by frame. Particularly, MoG model in each frame is regularized by the learned foreground/background knowledge in previous frames. This makes online MoG model highly robust, stable and adaptive to practical foreground and background variations. The recommended model can be formulated as a brief probabilistic MAP model, which can be voluntarily solved by EM algorithm. We additional embed an affine transformation operator into the recommended model, which can be automatically accustomed to fit a wide range of video background transformations and make the method more robust to camera movements.