Target tracking with Bayesian fusion based template matching

In this paper a Bayesian fusion based template matching algorithm is proposed for the target tracking problem. Two different template matching methods (sum of the squared errors (SSE) and Gaussian mixture models (GMMs)) are weighted by their matching accuracies and then combined through the Bayesian theory to give a final robust template updating and matching. With the fusion of different template matching methods, the algorithm in this paper can deal with the problem such as the template drifting, shape deformation or occluded object matching.