ETRI-Knowledge Sharing Plaform



논문 검색
구분 SCI
연도 ~ 키워드


학술대회 Human Internal State Recognition Using Modified Chain-Code Method
Cited 1 time in scopus Download 0 time Share share facebook twitter linkedin kakaostory
이대하, 박천수, 장민수, 김재홍
International Conference on Ubiquitous Robots and Ambient Intelligence (URAI) 2015, pp.112-113
15PC2800, 인간 친화적 로봇 서비스 환경에서 판단 적합성 90%이상인 복합지식 기반 판단 및 의미기반 로봇 표현 기술 개발, 김재홍
Through facial expressions, it is able to read the human emotions. In order to achieve a smooth interaction between human and machines, the importance of emotion recognition is increasing. In this paper, we do handle human internal states except the existing basic six human emotions. To indicate a change of the feature points of face, we use vector chain code encoding method, it is extended from the existing chain code method. The chain code only encode the direction information between features. The vector chain code can encode the direction information and distance information altogether. Through experiments, a method utilizing a vector chain code that presented in this paper was more improved rather than the method using the conventional chain code.
KSP 제안 키워드
Chain code method, Emotion recognition, Encoding method, Human Emotions, Internal state recognition, Vector chain code, direction information, distance information, facial expression, feature points