Abstract
Shape-based regularization has proven to be a useful method for delineating objects within noisy images where one has prior knowledge of the shape of the targeted object. When a collection of possible shapes is available, the specification of a shape prior using kernel density estimation is a natural technique. Unfortunately, energy functionals arising from kernel density estimation are of a form that makes them impossible to directly minimize using efficient optimization algorithms such as graph cuts. Our main contribution is to show how one may recast the energy functional into a form that is minimizable iteratively and efficiently using graph cuts.
Similar content being viewed by others
References
Belongie, S., Malik, J., Puzicha, J.: Shape matching and object recognition using shape contexts. IEEE Trans. Pattern Anal. Mach. Intell. 24(4), 509–522 (2002)
Boykov, Y., Jolly, M.: Interactive graph cuts for optimal boundary & region segmentation of objects in nd images. In: Proceedings. Eighth IEEE International Conference on Computer Vision, 2001. ICCV 2001, vol. 1, pp. 105–112. IEEE Press, New York (2001)
Boykov, Y., Kolmogorov, V.: An experimental comparison of min-cut/max-flow algorithms for energy minimization in vision. IEEE Trans. Pattern Anal. Mach. Intell. 26(9), 1124–1137 (2004)
Boykov, Y., Veksler, O., Zabih, R.: Fast approximate energy minimization via graph cuts. IEEE Trans. Pattern Anal. Mach. Intell. 23(11), 1222–1239 (2001)
Chang, J., Chou, T., Brennan, K.: Tracking monotonically advancing boundaries in image sequences using graph cuts and recursive kernel shape priors. IEEE Trans. Med. Imaging 13(5), 1008–1020 (2012)
Cremers, D., Grady, L.: Statistical priors for efficient combinatorial optimization via graph cuts. In: Computer Vision–ECCV 2006, pp. 263–274 (2006)
Cremers, D., Soatto, S.: A pseudo-distance for shape priors in level set segmentation. In: 2nd IEEE Workshop on Variational, Geometric and Level Set Methods in Computer Vision, pp. 169–176 (2003). Citeseer
Cremers, D., Osher, S., Soatto, S.: Kernel density estimation and intrinsic alignment for shape priors in level set segmentation. Int. J. Comput. Vis. 69(3), 335–351 (2006)
Cremers, D., Schmidt, F., Barthel, F.: Shape priors in variational image segmentation: convexity, Lipschitz continuity and globally optimal solutions. In: IEEE Conference on Computer Vision and Pattern Recognition 2008. CVPR 2008, pp. 1–6. IEEE Press, New York (2008)
Dambreville, S., Rathi, Y., Tannenbaum, A.: A framework for image segmentation using shape models and kernel space shape priors. IEEE Trans. Pattern Anal. Mach. Intell. 30(8), 1385–1399 (2008)
Das, P., Veksler, O., Zavadsky, V., Boykov, Y.: Semiautomatic segmentation with compact shape prior. Image Vis. Comput. 27(1–2), 206–219 (2009)
El Zehiry, N., Xu, S., Sahoo, P., Elmaghraby, A.: Graph cut optimization for the Mumford-Shah model. In: The Seventh IASTED International Conference on Visualization, Imaging and Image Processing, pp. 182–187. ACTA Press, Calgary (2007)
Freedman, D., Drineas, P.: Energy minimization via graph cuts: settling what is possible. In: IEEE Computer Society Conference on Computer Vision and Pattern Recognition, 2005. CVPR 2005, vol. 2, pp. 939–946. IEEE Press, New York (2005)
Freedman, D., Zhang, T.: Interactive graph cut based segmentation with shape priors. In: IEEE Computer Society Conference on Computer Vision and Pattern Recognition. CVPR 2005, vol. 1, pp. 755–762. IEEE Press, New York (2005)
Grady, L., Alvino, C.: The piecewise smooth Mumford–Shah functional on an arbitrary graph. IEEE Trans. Image Process. 18(11), 2547–2561 (2009)
Heimann, T., Meinzer, H.: Statistical shape models for 3D medical image segmentation: a review. Med. Image Anal. 13(4), 543–563 (2009)
Hunter, D., Lange, K.: A tutorial on MM algorithms. Am. Stat. 58(1), 30–37 (2004)
Jiang, T., Tomasi, C.: Robust shape normalization based on implicit representations. In: 19th International Conference on Pattern Recognition, 2008. ICPR 2008, pp. 1–4. IEEE Press, New York (2008)
Kohli, P., Torr, P.H.: Dynamic graph cuts for efficient inference in Markov random fields. IEEE Trans. Pattern Anal. Mach. Intell. 29(12), 2079–2088 (2007)
Lee, S., Park, J.: Nonlinear shape normalization methods for the recognition of large-set handwritten characters. Pattern Recognit. 27(7), 895–902 (1994)
Lempitsky, V., Blake, A., Rother, C.: Image segmentation by branch-and-mincut. In: Computer Vision–ECCV 2008, pp. 15–29 (2008)
Lempitsky, V., Blake, A., Rother, C.: Branch-and-mincut: global optimization for image segmentation with high-level priors. J. Math. Imaging Vis. 44(3), 315–329 (2012)
Malcolm, J., Rathi, Y., Tannenbaum, A.: Graph cut segmentation with nonlinear shape priors. In: IEEE International Conference on Image Processing, 2007. ICIP 2007, vol. 4. IEEE Press, New York (2007)
Malcolm, J., Rathi, Y., Tannenbaum, A.: Multi-object tracking through clutter using graph cuts. In: IEEE 11th International Conference on Computer Vision, 2007. ICCV 2007, pp. 1–5. IEEE Press, New York (2007)
O’Hagan, A., Forster, J., Kendall, M.: Bayesian Inference. Arnold, Sevenoaks (2004)
Pei, S.C., Lin, C.N.: Image normalization for pattern recognition. Image Vis. Comput. 13(10), 711–723 (1995)
Schindelin, J.: Fiji is just ImageJ—batteries included. In: Proceedings of the ImageJ User and Developer Conference, Luxembourg (2008)
Silverman, B.: Density Estimation for Statistics and Data Analysis vol. 26. Chapman & Hall, London (1986)
Slabaugh, G., Unal, G.: Graph cuts segmentation using an elliptical shape prior. In: IEEE International Conference on Image Processing. ICIP 2005, vol. 2. IEEE Press, New York (2005)
Tabbone, S., Wendling, L.: Binary shape normalization using the radon transform. In: Discrete Geometry for Computer Imagery, pp. 184–193. Springer, Berlin (2003)
Veksler, O.: Star shape prior for graph-cut image segmentation. In: Computer Vision–ECCV 2008, pp. 454–467 (2008)
Vineet, V., Narayanan, P.: Cuda cuts: fast graph cuts on the GPU. In: IEEE Computer Society Conference on Computer Vision and Pattern Recognition Workshops, CVPRW’08, pp. 1–8. IEEE Press, New York (2008)
Viola, P., Wells, W. III: Alignment by maximization of mutual information. Int. J. Comput. Vis. 24(2), 137–154 (1997)
Vu, N., Manjunath, B.: Shape prior segmentation of multiple objects with graph cuts. In: IEEE Conference on Computer Vision and Pattern Recognition, 2008. CVPR 2008, pp. 1–8. IEEE Press, New York (2008)
Zhu-Jacquot, J., Zabih, R.: Graph cuts segmentation with statistical shape priors for medical images. In: Third International IEEE Conference on Signal-Image Technologies and Internet-Based System, SITIS’07, pp. 631–635. IEEE Press, New York (2007)
Acknowledgements
This material is based upon work supported by the National Science Foundation under Agreement No. 0635561. J.C. and T.C. also acknowledge support from the National Science Foundation through grants DMS-1032131 and DMS-1021818, and from the Army Research Office through grant 58386MA.
Author information
Authors and Affiliations
Corresponding author
Appendices
Appendix A: Shape Alignment
In order to align the shape templates, we choose the affine transformation \(\mathbb{T}(\mathbf {s})=\alpha\mathbf{R}(\boldsymbol{\omega}) (\mathbf {s}-\mathbf {c})\) that minimizes the shape energy
In this expression, we have rewritten the contour integral on ∂Ω in terms of an integral over the zero-level set of ϕ Ω . Here, we develop a local Newton-Raphson algorithm for finding the optimal transform. Let us denote the 2⋅d column vector of transformation parameters φ=[α c ω]. We estimate φ using the iterative updates
where
and
To populate the matrix, we need to calculate the gradients with respect to α, c, and ω. For λ>0, the first-order derivatives take the form
The sign function comes about from differentiation of the absolute value function in the distributional sense.
For λ≠1, the second-order derivatives constitute tensors that take the form
If λ=1, the first term is zero and the following term is added
The contour integrals in these expressions can be calculated using a regularized version of the Dirac delta function such as
Similarly, the characteristic function can be interpreted as the Heaviside function acting on the signed-distance shape embedding, which can be approximated using an approximation of the Heaviside function such as
The nonzero transformation derivatives are as follows
The results of this alignment method are shown in the third column Fig. 3. As an example, the raw input templates for the vans in this figure are shown in Fig. 5.
Appendix B: MM Algorithm for Iterative Graph Cuts
The shape term log∑w j K(Ω,Ω j ) can make minimization of the energy difficult, since its formulation involves a sum within a logarithm, making the energy functional nonlinear with respect to the labeling of pixels (into background and foreground) in the image.
To linearize the shape contribution, we will derive a surrogate function with separated terms. A function f(x|x k ), with fixed x k , is said to majorize a function g(x) at x k if the following holds [17]
We wish to perform iterative inference by finding a sequence of segmentations Ω (n+1)=argmin Ω Q(Ω|Ω (n)), where Q(Ω|Ω (n)) majorizes Eq. (7). By the descent property of the MM algorithm [17], this sequence converges to a local minimum.
For any convex function f(x), the following holds [17]
Noting that −log(⋅) is convex, we have
verifying that the inequality condition (Eq. (16)) holds. In the case that Ω=Ω (n), we have
verifying that the equality condition (Eq. (17)) holds. So the two majorizing conditions (16), (17) are met, and we can minimize our original energy by iteratively minimizing
Because the distance function can be written as a sum over vertices and edges of a graph, so can Eq. (19). As a result, it is possible to minimize Eq. (19) within the graph cuts framework.
Rights and permissions
About this article
Cite this article
Chang, J.C., Chou, T. Iterative Graph Cuts for Image Segmentation with a Nonlinear Statistical Shape Prior. J Math Imaging Vis 49, 87–97 (2014). https://doi.org/10.1007/s10851-013-0440-9
Published:
Issue Date:
DOI: https://doi.org/10.1007/s10851-013-0440-9