Design of a real-time gesture recognition system: high performance through algorithms and software

The embedded systems group at Princeton University in New Jersey has developed, as an example of smart cameras, a gesture recognition system that can build a complete model of the torso and recognize gestures at 30 frames/s. Designing a real-time gesture recognition system is a complex task that involves many issues such as algorithm design, processing speed, system architecture, and video interface. In this article, the authors describe a method to manage the complexity by decomposing the entire process into different design and implementation phases.