Globally Optimal Regions and Boundaries as Minimum Ratio Weight Cycles

We describe a new form of energy functional for the modeling and identification of regions in images. The energy is defined on the space of boundaries in the image domain and can incorporate very general combinations of modeling information both from the boundary (intensity gradients, etc.) and from the interior of the region (texture, homogeneity, etc.). We describe two polynomial-time digraph algorithms for finding the global minima of this energy. One of the algorithms is completely general, minimizing the functional for any choice of modeling information. It runs in a few seconds on a 256/spl times/256 image. The other algorithm applies to a subclass of functionals, but has the advantage of being extremely parallelizable. Neither algorithm requires initialization.

[1]  G. Dantzig,et al.  FINDING A CYCLE IN A GRAPH WITH MINIMUM COST TO TIME RATIO WITH APPLICATION TO A SHIP ROUTING PROBLEM , 1966 .

[2]  Ugo Montanari,et al.  On the optimal detection of curves in noisy pictures , 1971, CACM.

[3]  Nimrod Megiddo,et al.  Combinatorial optimization with rational objective functions , 1978, Math. Oper. Res..

[4]  Richard M. Karp,et al.  A characterization of the minimum cycle mean in a digraph , 1978, Discret. Math..

[5]  G. Kanizsa,et al.  Organization in Vision: Essays on Gestalt Perception , 1979 .

[6]  Nimrod Megiddo.  Combinatorial Optimization with Rational Objective Functions , 1979, Math. Oper. Res..

[7]  S. K. Gupta,et al.  Combinatorial Optimization with Rational Objective Functions: A Communication , 1983, Math. Oper. Res..

[8]  Andrew Blake,et al.  Visual Reconstruction , 1987, Deep Learning for EEG-Based Brain–Computer Interfaces.

[9]  Terry E. Weymouth,et al.  Using Dynamic Programming For Minimizing The Energy Of Active Contours In The Presence Of Hard Constraints , 1988, [1988 Proceedings] Second International Conference on Computer Vision.

[10]  Shimon Ullman,et al.  Structural Saliency: The Detection Of Globally Salient Structures using A Locally Connected Network , 1988, [1988 Proceedings] Second International Conference on Computer Vision.

[11]  Steven W. Zucker,et al.  Radial Projection: An Efficient Update Rule for Relaxation Labeling , 1989, IEEE Trans. Pattern Anal. Mach. Intell..

[12]  Steven W. Zucker,et al.  Trace Inference, Curvature Consistency, and Curve Detection , 1989, IEEE Trans. Pattern Anal. Mach. Intell..

[13]  G. Sperling.  Three stages and two systems of visual processing. , 1989, Spatial vision.

[14]  Ravindra K. Ahuja,et al.  Network Flows: Theory, Algorithms, and Applications , 1993 .

[15]  Gérard G. Medioni,et al.  Inferring global perceptual contours from local features , 1993, Proceedings of IEEE Conference on Computer Vision and Pattern Recognition.

[16]  James Elder,et al.  The effect of contour closure on the rapid discrimination of two-dimensional shapes , 1993, Vision Research.

[17]  I Kovács,et al.  A closed curve is much more than an incomplete one: effect of closure in figure-ground segmentation. , 1993, Proceedings of the National Academy of Sciences of the United States of America.

[18]  James Elder,et al.  A measure of closure , 1994, Vision Research.

[19]  Philip N. Klein,et al.  Faster Shortest-Path Algorithms for Planar Graphs , 1997, J. Comput. Syst. Sci..

[20]  D. Mumford.  Elastica and Computer Vision , 1994 .

[21]  Alok Gupta,et al.  Dynamic Programming for Detecting, Tracking, and Matching Deformable Contours , 1995, IEEE Trans. Pattern Anal. Mach. Intell..

[22]  Steven W. Zucker,et al.  Computing Contour Closure , 1996, ECCV.

[23]  Lance R. Williams,et al.  Analytic solution of stochastic completion fields , 1995, Biological Cybernetics.

[24]  Ingemar J. Cox,et al.  "Ratio regions": a technique for image segmentation , 1996, Proceedings of 13th International Conference on Pattern Recognition.

[25]  Alan L. Yuille,et al.  Region Competition: Unifying Snakes, Region Growing, and Bayes/MDL for Multiband Image Segmentation , 1996, IEEE Trans. Pattern Anal. Mach. Intell..

[26]  Lance R. Williams,et al.  Stochastic Completion Fields: A Neural Model of Illusory Contour Shape and Salience , 1997, Neural Computation.

[27]  Philip N. Klein,et al.  Faster Shortest-Path Algorithms for Planar Graphs , 1997, J. Comput. Syst. Sci..

[28]  Jitendra Malik,et al.  Normalized cuts and image segmentation , 1997, Proceedings of IEEE Computer Society Conference on Computer Vision and Pattern Recognition.

[29]  P. Cavanagh,et al.  Position displacement, not velocity, is the cue to motion detection of second-order stimuli , 1998, Vision Research.

[30]  Jitendra Malik,et al.  Contour Continuity in Region Based Image Segmentation , 1998, ECCV.

[31]  Ian H. Jermyn,et al.  Globally optimal regions and boundaries , 1999, Proceedings of the Seventh IEEE International Conference on Computer Vision.

[32]  Lance R. Williams,et al.  Segmentation of salient closed contours from real images , 1999, Proceedings of the Seventh IEEE International Conference on Computer Vision.

[33]  Jitendra Malik,et al.  Normalized Cuts and Image Segmentation , 2000, IEEE Trans. Pattern Anal. Mach. Intell..

[34]  Ian H. Jermyn,et al.  Region extraction from multiple images , 2001, Proceedings Eighth IEEE International Conference on Computer Vision. ICCV 2001.