An Avatar-Based Weather Forecast Sign Language System for the Hearing-Impaired

Abstract : In this paper, we describe a text-to-animation framework for TV weather forecast sign language presentation. To this end, we analyzed the last three years’ weather forecast scripts to obtain the frequency of each word and determine the order of motion capture. About 500 sign language words were chosen and motion-captured for the weather forecast purpose, in addition to the existing 2,700 motions prebuilt for daily life. Words that are absent in the sign language dictionary are replaced with synonyms registered in KorLex, the Korean Wordnet, to improve the translation performance. The weather forecast with sign language is serviced via the Internet in an on-demand manner and can be viewed by PC or mobile devices.
Document type :
Conference papers
Complete list of metadatas

Cited literature [7 references]  Display  Hide  Download

https://hal.inria.fr/hal-01391353
Contributor : Hal Ifip <>
Submitted on : Thursday, November 3, 2016 - 11:05:59 AM
Last modification on : Friday, December 1, 2017 - 1:16:35 AM
Long-term archiving on : Saturday, February 4, 2017 - 1:04:39 PM

File

978-3-662-44654-6_51_Chapter.p...
Files produced by the author(s)

Licence


Distributed under a Creative Commons Attribution 4.0 International License

Identifiers

Citation

Juhyun Oh, Seonggyu Jeon, Minho Kim, Hyukchul Kwon, Iktae Kim. An Avatar-Based Weather Forecast Sign Language System for the Hearing-Impaired. 10th IFIP International Conference on Artificial Intelligence Applications and Innovations (AIAI), Sep 2014, Rhodes, Greece. pp.519-527, ⟨10.1007/978-3-662-44654-6_51⟩. ⟨hal-01391353⟩

Share

Metrics

Record views

70

Files downloads

154