2013 Fiscal Year Final Research Report

Estimation of human emotion by visible and invisible feature of facial image

Research Project

Project/Area Number	23500208
Research Category	Grant-in-Aid for Scientific Research (C)
Allocation Type	Multi-year Fund
Section	一般
Research Field	Perception information processing/Intelligent robotics
Research Institution	Japan Advanced Institute of Science and Technology
Principal Investigator	KOTANI Kazunori 北陸先端科学技術大学院大学, 情報科学研究科, 准教授 (20225452)
Project Period (FY)	2011 – 2013
Keywords	感情推定 / 顔表情 / 顔温度画像 / サーモグラフィ
Research Abstract	This research estimates human emotions by using facial expression (visible information) and a facial temperature picture (invisible information). For this goal, we created the database of facial expression image and corresponding facial temperature image in the basic human emotion classes. Then, we have developed an estimation of human emotion method which takes facial image, facial temperature image and combined both of them. As a result of evaluating the characteristics of our method, the accuracy of estimation is about 70% when only either was used. Furthermore, the accuracy improved about ten point by combining a both of them. By the consideration of experimental results, our approach is able to obtain the new possibility of estimation for feeling intensity and complicated feeling classes of human emotion.

(4 results)