Fourth IEEE International Conference on Automatic Face and Gesture Recognition (FG'00) A Method for Recognizing a Sequence of Sign Language Words Represented in a Japanese Sign Language Sentence Grenoble, France9 March 26-March 30 ISBN: 0-7695-0580-5
A JSL (Japanese Sign Language) sentence is represented by connecting several signed words continuously. And there are transitions between the signed words that have no meaning in the signed sentence. Therefore, to translate JSL into Japanese, first, it is necessary to detect each signed word from the inputted gesture of a JSL sentence with high accuracy. And then, a proper sequence of the recognized signed words must be generated. To achieve this, we have developed (1) a method for effectively detecting the borders of the signed words from ordinary signed gestures and segmenting them, (2) a method for detecting whether the signed gesture is represented by one hand or both hands, and (3) a method for distinguishing the segments representing the singed words from the segments representing the transitions. We have carried out an experiment with 200 samples of 10 JSL sentences to recognize the sequences of the signed words using the developed methods. As the result, the accuracy for the word is improved from 77.6% to 86.6%, and the accuracy for the sentence is improved from 46.0% to 58.0% by using the developed methods. These results indicate that the developed methods are effective.
Index Terms:
Japanese sign language (JSL), translation, gesture recognition, segmentation
Citation:
Hirohiko Sagawa, Masaru Takeuchi, "A Method for Recognizing a Sequence of Sign Language Words Represented in a Japanese Sign Language Sentence," fg, pp.434, Fourth IEEE International Conference on Automatic Face and Gesture Recognition (FG'00), 2000 Usage of this product signifies your acceptance of the Terms of Use. | |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Digg
Furl
Spurl
Blink
Simpy
Google
Del.icio.us
Y!MyWeb