Adaptive Middleware Architecture for a distributed omni-directional visual tracking system

In different areas of applications such as education, entertainment, medical surgery, or space shuttle launching, distributed visual tracking systems are of increasing importance. In this paper we describe the design, implementation and evaluation of OmniTrack, a distributed omni-directional visual tracking system, developed at the University of Illinois at Urbana-Champaign, with an Adaptive Middleware Architecture as the core of the system. With respect to both operating systems and network connections, adaptation is of fundamental importance to the tracking system, since it runs in an environment with large performance variations and without support of Quality of Service (QoS) guarantees. We present (1) The design of Agilos, an Adaptive Middleware Architecture that systematically supports application-aware quality adaptation; (2) The integration of offline profiles in order to assist the Agilos architecture and dynamically steer the adaptation path and behavior; (3) The design and experiments with OmniTrack, showing the viability of our approach. In such an application, we show that with the support of the Agilos architecture, tracking precision can be kept stable under dynamic variations of the environment such as the fluctuation of available network bandwidth, variations in CPU load, and dynamic changes in the location and speed of the tracked objects.