A SOMART system for gesture recognition

Mu Chun Su, Chao Hsin Hung, Yu Xiang Zhao

Research output: Contribution to journalArticlepeer-review

2 Scopus citations


Gestures are an efficient communication modality in many everyday situations. The desire to provide a more convenient and natural way of interacting with computes has led to considerable interest in recognizing hand gestures. In addition, one particular application of gesture-based systems is to implement a speaking aid for the deaf. In this paper, we present a SOMART system for the recognition of hand gestures. The sequence of a hand gesture is first projected into a 2-dimensional trajectory on a trained self-organizing feature map (SOM). Then the problem of recognizing hand gestures is transformed to the problem of recognizing 2-D trajectories. An ART-like algorithm is also proposed to generate multiple templates for each hand gesture. Finally, an unknown gesture is classified to be the gesture with the maximum similarity in the vocabulary via a template matching technique. A database consisted of 47 static hand gestures and 103 dynamic hand gestures was tested to demonstrate the performance of the proposed method.

Original languageEnglish
Pages (from-to)2764-2771
Number of pages8
JournalWSEAS Transactions on Computers
Issue number11
StatePublished - Nov 2006


  • Adaptive resonance theory (ART)
  • Character recognition
  • Dynamic gesture recognition
  • Self-organizing feature maps (SOM)
  • Sign language recognition


Dive into the research topics of 'A SOMART system for gesture recognition'. Together they form a unique fingerprint.

Cite this