2019 Fiscal Year Final Research Report

Development of Design Evaluation System for International Captions and Subtitles (DESICS) for barrier-free audiovisual presentations including films, movies, and TV programs

Research Project

PDF

Project/Area Number	16H03217
Research Category	Grant-in-Aid for Scientific Research (B)
Allocation Type	Single-year Grants
Section	一般
Research Field	Rehabilitation science/Welfare engineering
Research Institution	National Institute of Advanced Industrial Science and Technology
Principal Investigator	Ohyama Junji 国立研究開発法人産業技術総合研究所, 情報・人間工学領域, 主任研究員 (00635295)
Co-Investigator(Kenkyū-buntansha)	中島佐和子秋田大学, 理工学研究科, 講師 (40453542) 若月大輔筑波技術大学, 産業技術学部, 准教授 (50361887) 井野秀一国立研究開発法人産業技術総合研究所, 情報・人間工学領域, 総括研究主幹 (70250511)
Project Period (FY)	2016-04-01 – 2020-03-31
Keywords	認知支援工学 / バリアフリー / 映像認知 / アクセシブルデザイン / 時短デザイン / 情報保障 / 情報支援工学 / 福祉工学
Outline of Final Research Achievements	Captions and subtitles are essential to information accessibility for hearing-impaired people, deaf people, and elderly people who have difficulty hearing audio information to understand the information of video content. Besides, there are many cases in which captions are displayed in public spaces that cannot produce sound, and there is a high demand for people with normal hearing for subtitles. The Ministry of Internal Affairs and Communications supports the captioning rate, but the issue was to consider the "quality" of the caption design, such as the ease of viewing and comprehension. Therefore, in this study, we conducted a study of video cognitive characteristics to quantitatively investigate the effect of captioning design on the visibility and content understanding of video and captions not only in the captions themselves but also in the video. We have established a caption evaluation method based on cognitive characteristics and developed a captioning design evaluation tool.
Free Research Field	認知支援工学
Academic Significance and Societal Importance of the Research Achievements	これまで経験や勘に頼って決められていた字幕のわかりやすさの評価を、映像認知特性に基づいて科学的に定量的に分析する手法を確立し、かつ、その複雑な分析を自動化し、字幕の制作者や字幕の制作の依頼者が、完成した字幕が視聴者にわかりやすいかどうかを表示する字幕デザイン評価ツールを開発した。これによって、適切なコストと安定した品質の字幕の供給を支援し，バリアフリー化の普及を促進することができる。字幕やそれが重畳される映像がどう見えているか、を科学的に調べた知見は、スマートグラスのようなAR(Augmented Reality)技術などの次世代デザインの基盤知見としても学術的意義が大きい。