지원사업
학술연구/단체지원/교육 등 연구자 활동을 지속하도록 DBpia가 지원하고 있어요.
커뮤니티
연구자들이 자신의 연구와 전문성을 널리 알리고, 새로운 협력의 기회를 만들 수 있는 네트워킹 공간이에요.
논문 기본 정보
- 자료유형
- 학위논문
- 저자정보
- 지도교수
- 박신석, 김승종
- 발행연도
- 2013
- 저작권
- 고려대학교 논문은 저작권에 의해 보호받습니다.
이용수1
이 논문의 연구 히스토리 (2)
- 이 논문의 후속연구가 궁금하신가요?
- 연관 학술논문 또는 학술발표를 통해 보다 발전된 연구결과를 확인하실 수 있습니다.
- 이 논문의 연구 히스토리 확인하기
초록· 키워드
오류제보하기
Intelligent robots, which are able to communicate and interaction with human, are studied with development of robot technology. Imitating natural human motion and human-recognizable emotion expression of robot are required to communicate and interact with human effectively.
‘Automatic motion generation of face robot based on vision recognition’ is proposed for efficient generation of natural robot motion. Data, which is rotation angle of x/y/z/ axis and distance of face from camera and angle of both side of lip, is measured with non-uniformed sampling interval from webcam and commercial face recognition program. Automatic motion generation process with measured data is performed for generating motion appropriate for robot. First, noise in raw data is removed from filtering. Second, discontinuous movement is improved by turning point detection process. Lastly, motion of face elements, which are difficult or impossible to recognize, are generated another way based on neck rotation data.
Motion coloring based on robot emotion is proposed to express emotion of robot with motion and apply to reactive behavior model. In case of neck motion generated from the procedures above, motion has been modified with restriction based on emotion of robot. In case of emotion expression with face elements, random motion has been generated in limited driving range of each emotion.
By making use of proposed method, motion appropriate for robot is able to be generated automatically and emotion expression of robot with generated motion is possible.
‘Automatic motion generation of face robot based on vision recognition’ is proposed for efficient generation of natural robot motion. Data, which is rotation angle of x/y/z/ axis and distance of face from camera and angle of both side of lip, is measured with non-uniformed sampling interval from webcam and commercial face recognition program. Automatic motion generation process with measured data is performed for generating motion appropriate for robot. First, noise in raw data is removed from filtering. Second, discontinuous movement is improved by turning point detection process. Lastly, motion of face elements, which are difficult or impossible to recognize, are generated another way based on neck rotation data.
Motion coloring based on robot emotion is proposed to express emotion of robot with motion and apply to reactive behavior model. In case of neck motion generated from the procedures above, motion has been modified with restriction based on emotion of robot. In case of emotion expression with face elements, random motion has been generated in limited driving range of each emotion.
By making use of proposed method, motion appropriate for robot is able to be generated automatically and emotion expression of robot with generated motion is possible.
목차
Abstract i목 차 ii그림목차 iii표 목 차 iv제 1 장 서 론 11.1 연구 배경 11.2 연구 동향 31.3 연구 목적 및 내용 7제 2 장 비전 인식 기반 모션 자동 생성 82.1 비전 인식 데이터 획득 82.2 모션 자동 생성 프로세스 112.2.1 리샘플링 및 필터링 112.2.2 전환점 검출 142.3 얼굴 요소 모션 생성 19제 3 장 로봇 감정 기반 모션 컬러링 203.1 감정 별 표정 생성 213.2 감정 별 모션 컬러링 233.2.1 감정 별 구동 범위 및 속도 제한 243.2.2 모션 컬러링 27제 4 장 실험 및 결과 304.1 시스템 구현 304.2 비전 인식 데이터를 이용한 모션 자동 생성 결과 324.3 감정 기반 모션 컬러링 결과 41제 5 장 결 론 46참 고 문 헌 48
최근 본 자료
댓글(0)
0