Spotting dynamic hand gestures in video image sequences using hidden Markov models

Author

Morguet, Peter ; Lang, Manfred

Author_Institution

Inst. for Human-Machine-Commun., Munich Univ. of Technol., Germany

fYear

1998

fDate

4-7 Oct 1998

Firstpage

193

Abstract

A new and general stochastic approach to find and identify dynamic gestures in continuous video streams is presented. Hidden Markov models (HMMs) are used to solve this combined problem of temporal segmentation and classification in an integral way. Basically, an improved normalized Viterbi algorithm allows one to continuously observe the output scores of the HMMs at every time step. Characteristic peaks in the output scores of the respective models indicate the presence of gestures. Our experiments in the domain of hand gesture spotting provided excellent recognition results and very low temporal detection delays

Keywords

Viterbi detection; delays; feature extraction; gesture recognition; hidden Markov models; image classification; image segmentation; image sequences; stochastic processes; video signal processing; HMM; characteristic peaks; continuous video streams; dynamic gestures detection; dynamic gestures identification; dynamic hand gestures spotting; experiments; feature extraction; general stochastic approach; hidden Markov models; image classification; normalized Viterbi algorithm; output scores; recognition results; spatial segmentation; temporal detection delays; temporal segmentation; video image sequences; Delay; Hidden Markov models; Humans; Image recognition; Image segmentation; Image sequences; Shape; Stochastic processes; Streaming media; Viterbi algorithm;

fLanguage

English

Publisher

ieee

Conference_Titel

Image Processing, 1998. ICIP 98. Proceedings. 1998 International Conference on

Conference_Location

Chicago, IL

Print_ISBN

0-8186-8821-1

Type

conf

DOI

10.1109/ICIP.1998.999009

Filename

999009