Dynamic Foreground Detection and Tracking from Video using Markov Random Field

Robust foreground object segmentation via background modelling is a difficult problem in cluttered environments, where obtaining a clear view of the background to model is almost impossible. We propose a method capable of robustly estimating the background and detecting regions of interest in such environments. Most existing techniques thus adapt an iterative approach for foreground and background appearance modeling. However, these approaches may rely on good initialization and can be easily trapped in local optimal. In addition, they are usually time consuming for analyzing videos. To address these limitations, we propose an efficient appearance modeling technique for automatic primary video object segmentation in the MRF framework. We create an efficient initial foreground estimation. Then we use foreground-background labelling refinement. Finally, we can get the foreground from video faster than other approaches.