Title :
A new video segmentation approach: GrabCut in local window
Author :
Yang, Lei ; Guo, Yiming ; Wu, Xiaoyu ; Wang, Xiaomeng
Author_Institution :
Digital Media Dept., Commun. Univ. of China, Beijing, China
Abstract :
GrabCut is a classical efficient algorithm used in image segmentation. In this paper we provide an approach of video segmentation - an algorithm which is called “GrabCut in local window algorithm” based on GrabCut algorithm. First, make cutout with GrabCut in series of overlap local windows and get corresponding local contour boundaries. Then the contour boundaries in local windows are linked to get foreground contour boundary. Finally foreground segmentation result is obtained by filling the global contour boundary. The comparison with the original GrabCut algorithm is demonstrated at the end of this paper, proving the method is efficient.
Keywords :
image segmentation; video signal processing; GrabCut algorithm; image segmentation; local contour boundaries; local window algorithm; video segmentation; Algorithm design and analysis; Data models; Educational institutions; Equations; Image color analysis; Image segmentation; Minimization; Dilation; Erosion; GrabCut; Local Window; Video Segment;
Conference_Titel :
Soft Computing and Pattern Recognition (SoCPaR), 2011 International Conference of
Conference_Location :
Dalian
Print_ISBN :
978-1-4577-1195-4
DOI :
10.1109/SoCPaR.2011.6089280