Sign Language Recognizer Using HMMs

Title

Sign Language Recognizer Using HMMs

Creator

Middi V.S.R.; Raju M.A.

Description

In our day to day lives, we come across especially abled people who perform their daily chores with the aid of motivation that they get from self-confidence. There are many with hearing impairment. Sign language is the most expressed and natural way for them to communicate. Some chains of restaurants have, in fact, recruited deaf servers providing them with employment opportunities. Therefore, automatic Sign language recognition has become the crux of vision research. This paper is based on a project that builds a system that can recognize words communicated using the American Sign Language (ASL). Having been provided with a preprocessed dataset of tracked hand and nose positions extracted from the video, the set of Hidden Markov Models are trained. Using a part of this dataset, identification of individual words from test sequences is done. It provides them with the ability to communicate better, opening up a lot of opportunities. 2021, The Editor(s) (if applicable) and The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd.

Source

Smart Innovation, Systems and Technologies, Vol-195, pp. 715-724.

Date

2021-01-01

Publisher

Springer Science and Business Media Deutschland GmbH

Subject

Bayes nets; Data; Data; Features; Fir Hidden Markov Models; Jupyter; Model selection; Recognizer component; Testing; Training

Coverage

Middi V.S.R., Brown University, Providence, 02916, RI, United States; Raju M.A., Christ University, Bangalore, Karnataka, India

Rights

Restricted Access

Relation

ISSN: 21903018; ISBN: 978-981157077-3

Format

Online

Language

English

Type

Conference paper

Identifier

https://doi.org/10.1007/978-981-15-7078-0_71

https://www.scopus.com/inward/record.uri?eid=2-s2.0-85096491592&doi=10.1007%2f978-981-15-7078-0_71&partnerID=40&md5=a90451c382357fad884e96ee34997664