Summary: Object segmentation using graph cuts based active contours
Ning Xu a,*, Narendra Ahuja b
, Ravi Bansal c
DMS Lab, Samsung Information Systems America, 3345 Michelson Dr., Suite 250, Irvine, CA 92612, USA
ECE Department, University of Illinois at Urbana-Champaign, Urbana, IL 61801, USA
Department of Psychiatry, Columbia University, New York, NY, USA
Received 16 June 2005; accepted 16 November 2006
Available online 16 January 2007
In this paper we present a graph cuts based active contours (GCBAC) approach to object segmentation. GCBAC approach is a com-
bination of the iterative deformation idea of active contours and the optimization tool of graph cuts. It differs from traditional active
contours in that it uses graph cuts to iteratively deform the contour and its cost function is defined as the summation of edge weights
on the cut. The resulting contour at each iteration is the global optimum within a contour neighborhood (CN) of the previous result.
Since this iterative algorithm is shown to converge, the final contour is the global optimum within its own CN. The use of contour neigh-
borhood alleviates the well-known bias of the minimum cut in favor of a shorter boundary. GCBAC approach easily extends to the seg-
mentation of three and higher dimensional objects, and is suitable for interactive correction. Experimental results on selected data sets
and performance analysis are provided.