Target and Background Area Weighted Tracking Algorithm

Recently, kernel-based tracking algorithms such as the mean shift tracking algorithm has been proposed. However, there exists an inherent instability problem which is due to the use of an isotropic kernel for spatiality. In this paper, In this paper, we propose a new tracking algorithm : the weighted mean of the centroids corresponding to each color bin of the target. The weight is determined by the area of the color bins and background colors. The tracking based on proposed model contains spatial information on the distribution of the colors, is rather insensitive to the loss of pixels and change in the number of pixels, and takes the colors into account according to the area they cover in the initial target region. Due to these properties, it possible to track the target in difficult conditions such as low-frame-rate environment, severe partial occlusion and partial color change environment. Furthermore, the target position estimation is executed in a one step computation, which makes the algorithm fast. We compare the stableness of the proposed tracking scheme with the conventional mean shift algorithm experimentally.