Collaborative and Reconfigurable Object Tracking

摘要

Many vision applications perform intensive computations and demand hardware implementations in order to exhibit real time performance. Furthermore, these applications are hard to pre-characterize and can take different paths according to events happening in the scene at runtime. Hence, reconfigurable hardware devices are envisioned as the proper platform to implement such applications, providing both real time performance and dynamic adaptability for the system. In this paper, we present a collaborative and reconfigurable object tracking system that has been built as part of our work. We justify the need for dynamic adaptation of the system through scenarios and examples. Experimental results on a set of scenes advocate the fact that our system works effectively for different scenes and scenario of events through reconfiguration.