MEAN-SHIFT BACKGROUND IMAGE MODELLING (WP-P6)

Author(s) : Massimo Piccardi (University of Technology, Sydney, Australia) Tony Jan (University of Technology, Sydney, Australia)

Abstract : Background modelling is widely used in computer vision for the detection of foreground objects in a frame sequence. The more accurate the background model, the more correct turns out the detection of the foreground objects. In this paper, we presents an approach to background modelling based on a mean-shift procedure. The mean shift vector convergence properties enables the system to achieve reliable background modelling. In addition, histogram-based computation and the new concept of local basins of attraction allow us to to meet the stringent real-time requirements of video processing.

Menu