Title: Recognition of signed expressions using visually-oriented subunits obtained by an immune-based optimisation
Authors: Mariusz Oszust; Marian Wysocki
Addresses: Department of Computer and Control Engineering, Rzeszow University of Technology, 2 Wincentego Pola, 35-959 Rzeszow, Poland ' Department of Computer and Control Engineering, Rzeszow University of Technology, 2 Wincentego Pola, 35-959 Rzeszow, Poland
Abstract: This paper considers automatic visual recognition of signed expressions. The proposed method is based on modelling gestures with subunits, which is similar to modelling speech by means of phonemes. To define the subunits a data-driven procedure is applied. The procedure consists in partitioning time series, extracted from video, into subsequences which form homogeneous groups. The cut points are determined by an immune optimisation procedure based on quality assessment of the resulting clusters. In this paper the problem is formulated, its solution method is proposed and experimentally verified on a database of 101 Polish words and 35 sentences used at the doctor's and in the post office. The results show that our subunit-based classifier outperforms its whole-word-based counterpart, which is particularly evident when new expressions are recognised on the basis of a small number of examples.
Keywords: sign language recognition; time series segmentation; subsequences clustering; immune optimisation; computer vision; data mining; signed expressions; visual recognition; gesture modelling.
International Journal of Data Mining, Modelling and Management, 2014 Vol.6 No.2, pp.202 - 216
Received: 08 May 2021
Accepted: 12 May 2021
Published online: 06 Jul 2014 *