Particle filter tracking using color correlogram

Tao Zhang,Shumin Fei,Lili Wang
DOI: https://doi.org/10.3969/j.issn.1001-0505.2011.S1.029
2011-01-01
Abstract:To improve the accuracy of video target tracking, especially when the color of the target is similar to the background or the target gets orientation changes or partial occlusions, a novel color correlogram based particle filter is proposed for the object tracking in visual surveillance. By using the color correlogram as object feature, spatial information is incorporated into object representation, which yields a reliable likelihood description of the observation and prediction for tracking the objects accurately. The capability of the tracker to tolerate appearance changes like orientation changes, small scale changes, partial occlusions and background scene changes is demonstrated using real image sequences. Experimental evidence shows that the color correlogram is more effective than the traditional color histogram for objects tracking.
What problem does this paper attempt to address?