Bayesian Network for multiple hypthesis tracking

For a flexible camera-to-camera tracking of multiple objects we model the objects behavior with a Bayesian network and combine it with the multiple hypohesis framework that associates observations with objects. Bayesian networks offer a possibility to factor complex, joint distributions into a product of intuitive conditional densities describing and predicting the objecs's path. Yet, these models do not distinquish unambiquously between the object's observations. The resulting uncertainty is explored by a multiple hypothesis evaluation. The paper provides emperimental evidence of the performance of the proposed method.