Optical flow-based observation models for particle filter tracking
In this paper, we study the use of optical flow as a characteristic for tracking. We analyze the behavior of three flowbased observation models for particle filter algorithms, and compare the results with the ones obtained by a well-known, gradient based, observation model. In theory, optical flow could be used directly to displace an object model, but in practice flow estimation techniques lack the necessary accuracy. In view of the fact that probabilistic tracking algorithms enable imprecise or incomplete information to be handled naturally, these models have been used as a natural means of incorporating flow information into the tracking.