This paper presents a method for learning, tracking, and recognizing human gestures using a view-based approach to model both object and behavior. Object views are represented using sets of view models, rather than single templates. Stereotypical space-time patterns, i.e. gestures, are then matched to stored gesture patterns using dynamic time warping. Real-time performance is achieved by using special-purpose correlation hardware and view prediction to prune as much of the search space as possible. Both view models and view predictions are learned from examples. We present results showing tracking and recognition of human hand gestures at over 10Hz.