Automatic gait recognition by symmetry analysis

Abstract

This work describes a new method for automatic gait recognition based on analysing the symmetry of human motion, by extending an established symmetry operator. This operator, rather than relying on the borders of a shape or on general appearance, is able to locate features by their symmetrical properties. Essentially, it accumulates the symmetries between image points to give a symmetry map. This approach is reinforced by the psychologists' view that gait is a symmetrical pattern of motion. It is also supported by works that suggest pendular motion is an appropriate model for automatic gait recognition.

This research is the first application of symmetry to images of moving objects. We have developed approaches to temporal symmetry and refined earlier approaches not only in terms of temporal issues but also in terms of basic capability. Accordingly, the new approaches generate symmetry maps of moving subjects by extension and refinement of the earlier operator to include time. As such, the resulting maps obtain information concerning not only body shape, but also the way it moves. The Fourier transform is used to derive the gait signatures from the symmetry maps, in view of its invariance and coding properties, and also its descriptive capability.

More information

Identifiers

Catalogue record

Export record

ASCII CitationAtomBibTeXData Cite XMLDublin CoreDublin CoreEP3 XMLEndNoteHTML CitationHTML CitationHTML ListJSONMETSMODSMPEG-21 DIDLOpenURL ContextObjectOpenURL ContextObject in SpanRDF+N-TriplesRDF+N3RDF+XMLRIOXX2 XMLReferReference ManagerSimple Metadata