Hidden markov models for greek sign language recognition
Vasileia Pashaloudi, Konstantinos G. Margaritis · 2002
Abstract:- Sign languages are the basic means of communication between hearing impaired people. A translator is usually needed when a deaf person wants to communicate with persons that do not speak sign language. The work presented in this paper aims at developing a system that could recognize and translate isolated and continuous Greek Sign Language (GSL) sentences. Input is obtained by using a feature extracting method from two-dimensional images. The feature vectors produced by the method are then presented as input to Hidden Markov Models (HMMs) for recognition. We use a vocabulary of 26 greek words (nouns, pronouns, adjectives, verbs) and achieve a recognition rate of 98 % for isolated recognition and 85.7 % for continuous recognition. Key-Words:- Hidden Markov Models, sign language, sign language recognition, gesture recognition. 1