| Project/Area Number |
18K11430
|
|---|
| Research Category |
Grant-in-Aid for Scientific Research (C)
|
| Allocation Type | Multi-year Fund |
|---|
| Section | 一般 |
|---|
| Review Section |
Basic Section 61030:Intelligent informatics-related
|
|---|
| Research Institution | Gifu University |
|---|
Principal Investigator |
|
|---|
| Co-Investigator(Kenkyū-buntansha) |
加藤 三保子 豊橋技術科学大学, 総合教育院, 特任教授 (30194856)
|
|---|
| Project Period (FY) |
2018-04-01 – 2021-03-31
|
| Project Status |
Completed (Fiscal Year 2020)
|
| Budget Amount *help |
¥3,120,000 (Direct Cost: ¥2,400,000、Indirect Cost: ¥720,000)
Fiscal Year 2020: ¥780,000 (Direct Cost: ¥600,000、Indirect Cost: ¥180,000)
Fiscal Year 2019: ¥910,000 (Direct Cost: ¥700,000、Indirect Cost: ¥210,000)
Fiscal Year 2018: ¥1,430,000 (Direct Cost: ¥1,100,000、Indirect Cost: ¥330,000)
|
|---|
| Keywords | 手話表記法 / SignWriting / 手話単語 / 単語ベクトル / 表記揺れ / 辞書検索 / 認識 / 手話単語識別 / 機械学習 / 単語認識 / 手話動作 / 単語識別 / 動作特徴ベクトル / 手話文字認識 / 手話文書 |
|---|
| Outline of Final Research Achievements |
SignWriting (SW) is a system for writing and reading sign languages without translating them into spoken languages in daily life. Signs are written in SW intuitively and readably with iconic symbols (characters) that represent hand shapes, movements, facial expressions, etc. However, due to its wide variety of spelling, searching signs in sign language text or dictionaries cannot be implemented by simple string comparison. In this study, we investigated vector representations of SW symbols, signs, and movement components of signs, and proposed methods to recognize the written signs based on the similarities of those vectors. 406 dictation data of 30 signs were used to consider the sign similarities. The effectiveness of the methods was verified by experiments using the dictation data by searching for the written signs in our JSL-Japanese dictionary.
|
| Academic Significance and Societal Importance of the Research Achievements |
手話には広く一般に使われる文字がないため，手話文の記録・伝達は一般に動画像の形，あるいは，日本語などに翻訳した文書の形で行われる．日本語も音声の形で記録・伝達できるが，文字と音声の役割は同じではない．日本語に翻訳する場合は翻訳の手間やニュアンスの違い，正確さの問題がある．SignWritingは手話言語を直接読み書きする方法の一つであるが，表記の揺れが大きく，文書に対する基本操作である単語の同定・検索が難しいという問題がある．本研究では表記揺れに柔軟に対応する手法を提案することで，手話言語の文書化・文字表現の有用性・利便性の向上を図った．
|
