Abstract

Innovative analysis methods applied to data extracted by o®-the-shelf peripherals can provideuseful results in activity recognition without requiring large computational resources. In thispaper a framework is proposed for automated posture and gesture recognition, exploiting depthdata provided by a commercial tracking device. The detection problem is handled as a semanticbasedresource discovery. A general data model and the corresponding ontology provide theformal underpinning for posture and gesture annotation via standard Semantic Web languages.Hence, a logic-based matchmaking, exploiting non-standard inference services, allows to: (i) detectpostures via on-the-°y comparison of the annotations with standard posture descriptionsstored as instances of a proper Knowledge Base; (ii) compare subsequent postures in order torecognize gestures. The framework has been implemented in a prototypical tool and experimentaltests have been carried out on a reference dataset. Preliminary results indicate the feasibility of theproposed approach.

Autore Pugliese

• Di Summa Maria

Tutti gli autori

• M. Ruta; F. Scioscia; M. di Summa; S. Ieva; E. Di Sciascio; M. Sacco

Titolo volume/Rivista

International Journal of Semantic Computing

Anno di pubblicazione

2014

ISSN

1793-7108

ISBN

Non Disponibile

Numero di citazioni Wos

Nessuna citazione

Ultimo Aggiornamento Citazioni

Non Disponibile

Numero di citazioni Scopus

Non Disponibile

Ultimo Aggiornamento Citazioni

Non Disponibile

Settori ERC

Non Disponibile

Codici ASJC

Non Disponibile