Reach out and touch space (motion learning)

We propose a method for learning models of human motion from a coarsely sampled set of examples. The models we synthesize may be used to generate plausible motions from a high level description consisting of start and stop positions, style, mood, age, etc. In the field of computer vision, such models can be useful for human body motion tracking/estimation and gesture recognition. The models can also be used to generate arbitrary realistic human motion, and may be of help in trying to understand the mechanisms behind the perception of biological motion by the human visual system. Experimental results of the learning technique applied to reaching and drawing motions are presented.