For the purposes of foreground estimation, the true background model is unavailable in many practical circumstances and needs\r\nto be estimated from cluttered image sequences. We propose a sequential technique for static background estimation in such\r\nconditions, with low computational and memory requirements. Image sequences are analysed on a block-by-block basis. For each\r\nblock location a representative set is maintained which contains distinct blocks obtained along its temporal line. The background\r\nestimation is carried out in a Markov Random Field framework, where the optimal labelling solution is computed using iterated\r\nconditional modes. The clique potentials are computed based on the combined frequency response of the candidate block and its\r\nneighbourhood. It is assumed that the most appropriate block results in the smoothest response, indirectly enforcing the spatial\r\ncontinuity of structures within a scene. Experiments on real-life surveillance videos demonstrate that the proposedmethod obtains\r\nconsiderably better background estimates (both qualitatively and quantitatively) than median filtering and the recently proposed\r\nââ?¬Å?intervals of stable intensityââ?¬Â method. Further experiments on theWallflower dataset suggest that the combination of the proposed\r\nmethod with a foreground segmentation algorithm results in improved foreground segmentation.
Loading....