Skip to Main Content
Tracking objects is an important activity in many applications such as Augmented Reality, Visual Servoing, and Tangible Interfaces. Most of these applications are inherently dynamic and demand realtime response while tracking. Also, due to the issues of economics, such applications can benefit from tracking systems that do not depend on well-engineered setups and instead use inexpensive sensors for tracking. These requirements make distributed tracking a challenging problem that needs a thorough investigation. In this paper an approach to tracking, which uses the concepts of dynamic discovery, in a distributed environment containing autonomous vision-based trackers that use inexpensive sensors (such as Web cameras) is presented along with its empirical validation. The empirical results obtained from experiments are promising and validate the premise that Distributed Tracking Systems can be built using inexpensive vision-based sensors and concepts of dynamic resource discovery.
Date of Conference: 8-10 Dec. 2008