Abstract

Background subtraction has been widely investigated in re- cent years. Most previous work has focused on stationary cameras. Re- cently, moving cameras have also been studied since videos from mobile devices have increased significantly. In this paper, we propose a uni- fied and robust framework to effectively handle diverse types of videos, e.g., videos from stationary or moving cameras. Our model is inspired by two observations: 1) background motion caused by orthographic cam- eras lies in a low rank subspace, and 2) pixels belonging to one tra jectory tend to group together. Based on these two observations, we introduce a new model using both low rank and group sparsity constraints. It is able to robustly decompose a motion tra jectory matrix into foreground and background ones. After obtaining foreground and background tra- jectories, the information gathered on them is used to build a statistical model to further label frames at the pixel level. Extensive experiments demonstrate very competitive performance on both synthetic data and real videos.