Multi-Cue Exemplar-Based Nonparametric Model for Gesture Recognition

This paper presents an approach for a multi-cue, viewbased recognition of gestures. We describe an exemplarbased technique that combines two different forms of exemplars shape exemplars and motion exemplars in a unified probabilistic framework. Each gesture is represented as a sequence of learned body poses as well as a sequence of learned motion parameters. The shape exemplars are comprised of pose contours, and the motion exemplars are represented as affine motion parameters extracted using a robust estimation approach. The probabilistic framework learns by employing a nonparametric estimation technique to model the exemplar distributions. It imposes temporal constraints between different exemplars through a learned Hidden Markov Model (HMM) for each gesture. We use the proposed multi-cue approach to recognize a set of fourteen gestures and contrast it against a shape only, singlecue based system.