Background subtraction with moving cameras via Bayesian low-rank estimation
Lyu Chengcheng, Lei Yu, Shihui Hu, Sun Hong · 2016
Background subtraction is a typical topic in the domain of video processing. In this paper, we propose to cope with background subtraction for separating moving objects from background even cameras are not fixed via modified Bayesian low-rank analysis. Particularly, the hierarchical Bayesian model is proposed for low-rank estimation where the perspective projection is exploited to compensate the camera moving. Extensive experiments show that the proposed method is comparable and outperforms the state-of-the-art.