Decomposition of human motion into dynamics-based primitives with application to drawing tasks

• Published in: Automatica (Oxford) Vol. 39; no. 12; pp. 2085 - 2098
• Main Authors: Del Vecchio, Domitilla, Murray, Richard M., Perona, Pietro
• Format: Journal Article
• Language: English
• Published: Oxford Elsevier Ltd 01.12.2003; Elsevier
• Subjects: Applied sciences; Classification; Computer experiments; Computer science; control theory; systems; Control system analysis; Control system synthesis; Control theory. Systems; Data acquisition; Exact sciences and technology; Learning theory; Parameter estimation; Process control. Computer integrated manufacturing; Signal segmentation; Computer experiments; Parameter estimation; Signal segmentation; Learning theory; Data acquisition; Classification; Learning; Experimental data; Segmentation; Computer; Signal analysis
• ISSN: 0005-1098; 1873-2836
• DOI: 10.1016/S0005-1098(03)00250-4

Using tools from dynamical systems and systems identification, we develop a framework for the study of primitives for human motion, which we refer to as movemes. The objective is understanding human motion by decomposing it into a sequence of elementary building blocks that belong to a known alphabet of dynamical systems. We develop a segmentation and classification algorithm in order to reduce a complex activity into the sequence of movemes that have generated it. We test our ideas on data sampled from five human subjects who were drawing figures using a computer mouse. Our experiments show that we are able to distinguish between movemes and recognize them even when they take place in activities containing an unspecified number of movemes.