MRF-based Background Initialisation for Improved Foreground Detection in Cluttered Surveillance Videos

TL;DR

This paper introduces a Markov Random Field-based background initialization method that enhances foreground detection in cluttered surveillance videos by considering spatial continuity, leading to improved object tracking accuracy.

Contribution

It extends existing background initialization techniques with an MRF-based approach, improving robustness in cluttered environments for better foreground segmentation.

Findings

Significant improvement in object tracking accuracy.

Outperforms Gaussian mixture models and feature histogram methods.

Effective in cluttered surveillance scenarios.

Abstract

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. In this paper, we propose a method capable of robustly estimating the background and detecting regions of interest in such environments. In particular, we propose to extend the background initialisation component of a recent patch-based foreground detection algorithm with an elaborate technique based on Markov Random Fields, where the optimal labelling solution is computed using iterated conditional modes. Rather than relying purely on local temporal statistics, the proposed technique takes into account the spatial continuity of the entire background. Experiments with several tracking algorithms on the CAVIAR dataset indicate that the proposed method leads to considerable improvements in object tracking accuracy, when compared to methods based on Gaussian mixture models and feature histograms.