Generalized background subtraction based on hybrid inference by belief propagation and bayesian filtering

We propose a novel background subtraction algorithm for the videos captured by a moving camera. In our tech-nique, foreground and background appearance models in each frame are constructed and propagated sequentially by Bayesian filtering. We estimate the posterior of appear-ance, which is computed by the product of the image likeli-hood in the current frame and the prior appearance prop-agated from the previous frame. The motion, which trans-fers the previous appearance models to the current frame, is estimated by nonparametric belief propagation; the ini-tial motion field is obtained by optical flow and noisy and incomplete motions are corrected effectively through the inference procedure. Our framework is represented by a graphical model, where the sequential inference of motion and appearance is performed by the combination of belief propagation and Bayesian filtering. We compare our algo-rithm with the existing state-of-the-art technique and evalu-ate its performance quantitatively and qualitatively in sev-eral challenging videos. 1