상세정보
학술대회
Human Segmentation based on GrabCut in Real-Time Video Sequences
- 발행일
- 201401
- 출처
- International Conference on Consumer Electronics (ICCE) 2014, pp.111-112
- 협약과제
- 13VS1100, 사람에 의한 안전위협의 실시간 인지를 위한 능동형 영상보안 서비스용 원거리 (CCTV 주간환경 5m이상) 사람 식별 및 검색 원천기술 개발, 유장희
- 초록
- In this paper, we present a fully-automatic human segmentation method without iteration in video sequences. To segment human body accurately, we adopt coarse-to-fine approach using human detection and background subtraction. HoG-based method is used to detect human ROI. Background subtraction is used to subtract subject image in human ROI and skeleton image. The human ROI, the subject image, and the skeleton image are initialization values of GrabCut. The initialization values provide more accurate foreground and background information to GrabCut. Therefore the proposed method can segment human silhouette accurately enough to apply in video analysis without iteration. Experimental results show that the proposed method can be achieved better performance than GrabCut in real-time video sequences. © 2014 IEEE.
- KSP 제안 키워드
- Background information, Background subtraction(BS), Human Segmentation, Human body, Human detection, Real-time video, Skeleton image, coarse-to-fine, segmentation method, video analysis, video sequences