Visual surveillance is an important computer vision research problem. As more and more surveillance cameras appear around us, the demand for automatic methods for video analysis is increasing. Such methods have broad applications including surveillance for safety in public transportation, public areas, and in schools and hospitals. Automatic surveillance is also essential in the fight against terrorism. In this light, the PETS 2006 data corpus contains seven left-luggage scenarios with increasing scene complexity. The challenge is to automatically determine when pieces of luggage have been abandoned by their owners using video data, and set an alarm. In this paper, we present a solution to this problem using a two-tiered approach. The first step is to track objects in the scene using a trans-dimensional Markov Chain Monte Carlo tracking model suited for use in generic blob tracking tasks. The tracker uses a single camera view, and it does not differentiate between people and luggage. The problem of determining if a luggage item is left unattended is solved by analyzing the output of the tracking system in a detection process. Our model was evaluated over the entire data set, and successfully detected the left-luggage in all but one of the seven scenarios.