Home // AMBIENT 2024, The Fourteenth International Conference on Ambient Computing, Applications, Services and Technologies // View article

Basic Investigation for Sign Language Sentence Interpretation Using Acceleration Sensor Information

Authors:
Hiroshi Tanaka
Moriyoshi Umeda
Yuusuke Kawakita
Hiromitsu Nishimura
Jin Mitsugi

Keywords: Sign language; Acceleration sensor; Segmentation; LSTM; SVM; Motion classification.

Abstract:
This paper presents a method for segmenting a sign language sentence consisting of multiple words into individual word motions, which is an elemental technique for achieving the final goal of interpreting sign language sentences. The authors propose a segmentation method based on the similarity of motions, focusing on the fact that the word motion is included in the sentence motion. We selected 22 frequently occurring sign words and created 5 short sentences using them and acquired word and short sentence motion data. The results of the segmentation method using these data are presented. In addition, we show the results of word classification and confirm the feasibility of the proposed method for sentence interpretation.

Pages: 1 to 5

Copyright: Copyright (c) IARIA, 2024

Publication date: September 29, 2024

Published in: conference

ISSN: 2326-9324

ISBN: 978-1-68558-185-5

Location: Venice, Italy

Dates: from September 29, 2024 to October 3, 2024