Development of effective modeling methods of cognitive images for multimodal information and its application to image coordination of space

2014 Fiscal Year Final Research Report

• Project/Area Number: 24650110
• Research Category: Grant-in-Aid for Challenging Exploratory Research
• Allocation Type: Multi-year Fund
• Research Field: Sensitivity informatics/Soft computing
• Research Institution: Chuo University
• Principal Investigator: KATO Toshikazu 中央大学, 理工学部, 教授 (50297107)
• Co-Investigator(Renkei-kenkyūsha): OGINO Akihiro 京都産業大学, コンピュータ理工学部, 准教授 (40407870) ; TADA Masahiro 近畿大学, 理工学部, 講師 (40418520)
• Project Period (FY): 2012-04-01 – 2015-03-31
• Keywords: 感性情報処理 / 感性データベース / 感性認知科学・感性心理学 / 感性脳科学 / ユビキタスセンシング

Outline of Final Research Achievements

We developed precise methods for modeling subjective images of multimedia and multimodal information based on the contrast and entropy among the components. We developed efficient modeling methods by automatic clustering and semi-automatic interpolation of learning examples from small size of seeds. We also developed estimating method of emotional states, such as comfortable, interested, boring, and concentrating, through ubiquitous sensing and analysis of user’s behavior log. For an application of the method, we developed a pilot system for coordinating total images of a space controlling multimedia environment.

Free Research Field

感性情報学