International Journal of Computer Vision

, Volume 69, Issue 3, pp 335–351 | Cite as

Kernel Density Estimation and Intrinsic Alignment for Shape Priors in Level Set Segmentation

  • Daniel Cremers
  • Stanley J. Osher
  • Stefano Soatto
Article
First Online:
Received:
Revised:
Accepted:

Abstract

In this paper, we make two contributions to the field of level set based image segmentation. Firstly, we propose shape dissimilarity measures on the space of level set functions which are analytically invariant under the action of certain transformation groups. The invariance is obtained by an intrinsic registration of the evolving level set function. In contrast to existing approaches to invariance in the level set framework, this closed-form solution removes the need to iteratively optimize explicit pose parameters. The resulting shape gradient is more accurate in that it takes into account the effect of boundary variation on the object’s pose.

Secondly, based on these invariant shape dissimilarity measures, we propose a statistical shape prior which allows to accurately encode multiple fairly distinct training shapes. This prior constitutes an extension of kernel density estimators to the level set domain. In contrast to the commonly employed Gaussian distribution, such nonparametric density estimators are suited to model aribtrary distributions.

We demonstrate the advantages of this multi-modal shape prior applied to the segmentation and tracking of a partially occluded walking person in a video sequence, and on the segmentation of the left ventricle in cardiac ultrasound images. We give quantitative results on segmentation accuracy and on the dependency of segmentation results on the number of training shapes.

Keywords

image segmentation shape priors level set methods Bayesian inference alignment kernel density estimation 
This is a preview of subscription content, log in to check access.

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

Supplementary material

Open image in new window

References

  1. Basri, R., Costa, L., Geiger, D., and Jacobs D. 1998. Determining the similarity of deformable shapes. Vision Research, 38:2365–2385.CrossRefGoogle Scholar
  2. Bresson, X., Vandergheynst, P., and Thirau, J.P. 2003. A priori information in image segmentation: Energy functional based on shape statistical model and image information. In Proc. IEEE Int. Conf. Image Processing, pp. 425–428.Google Scholar
  3. Brox, T. and Weickert, J. 2004. A TV flow based local scale measure for texture discrimination. In (eds.), T.Pajdla and V.Hlavac European Conf. on Computer Vision, volume 3022 of Lect. Not. Comp. Sci., Prague, Springer.Google Scholar
  4. Caselles, V., Kimmel, R., and Sapiro, G. 1995. Geodesic active contours. In Proc. IEEE Intl. Conf. on Comp. Vis., pages 694–699, Boston, USA pp. 694–699..Google Scholar
  5. Chan, T. and Zhu, W. 2003. Level set based shape prior segmentation. Technical Report 03-66, Computational Applied Mathematics, UCLA, Los Angeles.Google Scholar
  6. Chan, T.F. and Vese, L.A. 2001. Active contours without edges. IEEE Trans. Image Processing, 10(2):266–277.CrossRefMATHGoogle Scholar
  7. Charpiat, G., Faugeras, O., and Keriven, R. 2005. Approximations of shape metrics and application to shape warping and empirical shape statistics. Journal of Foundations Of Computational Mathematics, 5(1):1–58.MathSciNetCrossRefMATHGoogle Scholar
  8. Chen, Y., Tagare, H., Thiruvenkadam, S., Huang, F., Wilson, D., Gopinath, K.S., Briggs, R.W., and Geiser, E. 2002. Using shape priors in geometric active contours in a variational framework. Int.J.ofComputer Vision, 50(3):315–328.CrossRefMATHGoogle Scholar
  9. Chow, Y.S., Geman, S., and Wu, L.D. 1983. Consistent cross-validated density estimation. Annals of Statistics, 11:25–38.MathSciNetMATHGoogle Scholar
  10. Cootes, T.F., Taylor, C.J., Cooper, D.M., and Graham, J. 1995. Active shape models–their training and application. Comp.Vision Image Underst., 61(1):38–59.CrossRefGoogle Scholar
  11. Cootes T.F. and Taylor C.J. 1999. A mixture model for representing shape variation. Image and Vision Computing, 17(8):567–574.CrossRefGoogle Scholar
  12. Cremers, D. 2006. Dynamical statistical shape priors for level set based tracking. IEEE Trans. on Patt. Anal. and Mach. Intell.. To appear.Google Scholar
  13. Cremers, D., Kohlberger, T., and Schnörr, C. 2003. Shape statistics in kernel space for variational image segmentation. Pattern Recognition, 36(9):1929–1943.CrossRefMATHGoogle Scholar
  14. Cremers, D., Osher, S., and Soatto, S. 2004. Kernel density estimation and intrinsic alignment for knowledge-driven segmentation: Teaching level sets to walk. In Pattern Recogn., volume 3175 of Lect. Not. Comp. Sci., pp. 36–44. Springer, C.E. Rasmussen (ed.).Google Scholar
  15. Cremers, D. and Soatto, S. 2003. A pseudo-distance for shape priors in level set segmentation. In, IEEE 2nd Int. Workshop on Variational, Geometric and Level Set Methods, N.Paragios (eds.), Nice, pp. 169–176.Google Scholar
  16. Cremers, D. and Soatto, S. May 2005. Motion Competition: A variational framework for piecewise parametric motion segmentation. Int.J.ofComputer Vision, 62(3):249–265.CrossRefGoogle Scholar
  17. Cremers, D., Sochen, N., and Schnörr, C. 2003. Towards recognition-based variational segmentation using shape priors and dynamic labeling. In, Int. Conf. on Scale Space Theories in Computer Vision, volume 2695 of Lect. Not. Comp. Sci., L.Griffith, editor, pp. 388–400, Isle of Skye, Springer.Google Scholar
  18. Cremers, D., Sochen, N., and Schnörr, C. 2006. A multiphase dynamic labeling model for variational recognition-driven image segmentation. Int.J.ofComputer Vision, 66(1):67–81.CrossRefGoogle Scholar
  19. Cremers, D., Tischhäuser, F., Weickert, J., and Schnörr, C. 2002. Diffusion Snakes: Introducing statistical shape knowledge into the Mumford—Shah functional. Int.J.ofComputer Vision, 50(3):295–313.CrossRefMATHGoogle Scholar
  20. Deheuvels, P. 1977. Estimation non paramétrique de la densité par histogrammes généralisés. Revue de Statistique Appliquée, 25:5–42.MathSciNetGoogle Scholar
  21. Dervieux, A. and Thomasset, F. 1979. A finite element method for the simulation of Raleigh-Taylor instability. Springer Lect. Notes in Math., 771:145–158.MathSciNetCrossRefGoogle Scholar
  22. Devroye, L. and Györfi, L. 1985. Nonparametric Density Estimation. The L1 View. John Wiley, New York.Google Scholar
  23. Dryden, I.L. and Mardia, K.V. 1998. Statistical Shape Analysis. Wiley, Chichester.MATHGoogle Scholar
  24. Duin, R.P.W. 1976. On the choice of smoothing parameters for Parzen estimators of probability density functions. IEEE Trans. on Computers, 25:1175–1179.MATHMathSciNetGoogle Scholar
  25. Fréchet, M. 1961. Les courbes aléatoires. Bull. Inst. Internat. Stat., 38:499–504.MATHGoogle Scholar
  26. Gdalyahu, Y. and Weinshall, D. 1999. Flexible syntactic matching of curves and its application to automatic hierarchical classication of silhouettes. IEEE Trans. on Patt. Anal. and Mach. Intell., 21(12):1312–1328.CrossRefGoogle Scholar
  27. Grenander, U. 1976. Lectures in Pattern Theory. Springer, Berlin.MATHGoogle Scholar
  28. Grenander, U., Chow, Y., and Keenan, D.M. 1991. Hands: A Pattern Theoretic Study of Biological Shapes. Springer, New York.Google Scholar
  29. Heiler, M. and Schnörr, C. 2003. Natural image statistics for natural image segmentation. In IEEE Int. Conf. on Computer Vision, pp. 1259–1266.Google Scholar
  30. Kendall, D.G. 1977. The diffusion of shape. Advances in Applied Probability, 9:428–430.CrossRefGoogle Scholar
  31. Kichenassamy, S., Kumar, A., Olver, P.J., Tannenbaum, A., and Yezzi, A.J. 1995. Gradient flows and geometric active contour models. In IEEE Int. Conf. on Computer Vision, pp. 810–815.Google Scholar
  32. Klassen, E., Srivastava, A., Mio, W., and Joshi, S.H. 2004. Analysis of planar shapes using geodesic paths on shape spaces. IEEE Trans. on Patt. Anal. and Mach. Intell., 26(3):372–383.CrossRefGoogle Scholar
  33. Leventon, M., Grimson, W., and Faugeras, O. 2000. Statistical shape influence in geodesic active contours. In CVPR, Hilton Head Island, SC.1:316–323.Google Scholar
  34. Malladi, R., Sethian, J.A., and Vemuri, B.C. 1995. Shape modeling with front propagation: A level set approach. IEEE Trans. on Patt. Anal. and Mach. Intell., 17(2):158–175.CrossRefGoogle Scholar
  35. Matheron, G. 1975. Random Sets and Integral Geometry. Wiley & Sons.Google Scholar
  36. Mio, W., Srivastava, A., and Liu, X. 2004. Learning and Bayesian shape extraction for object recognition. In European Conf. on Computer Vision, volume 3024 of Lect. Not. Comp. Sci., Prague, Springer, pp. 62–73.Google Scholar
  37. Moelich, M. and Chan, T. 2003. Tracking objects with the Chan-Vese algorithm. Technical Report 03-14, Computational Applied Mathematics, UCLA, Los Angeles.Google Scholar
  38. Mumford, D., and Shah, J. 1989. Optimal approximations by piecewise smooth functions and associated variational problems. Comm. Pure Appl. Math., 42:577–685.MathSciNetMATHGoogle Scholar
  39. Osher, S.J. and Fedkiw, R.P. 2002. Level Set Methods and Dynamic Implicit Surfaces. Springer, New York.Google Scholar
  40. Osher, S.J. and Paragios, N. 2003. Geometric Level Set Methods in Imaging, Vision and Graphics. Springer, Telos.MATHGoogle Scholar
  41. Osher, S.J. and Sethian, J.A. 1988. Fronts propagation with curvature dependent speed: Algorithms based on Hamilton–Jacobi formulations. J. of Comp. Phys., 79:12–49.MathSciNetCrossRefMATHGoogle Scholar
  42. Paragios, N. and Deriche, R. 2002. Geodesic active regions and level set methods for supervised texture segmentation. Int.J.ofComputer Vision, 46(3):223–247.CrossRefMATHGoogle Scholar
  43. Parzen, E. On the estimation of a probability density function and the mode. Annals of Mathematical Statistics, 33:1065–1076.Google Scholar
  44. Pons, J.-P., Hermosillo, G., Keriven, R., and Faugeras, O. 2003. How to deal with point correspondences and tangential velocities in the level set framework. In IEEE Int. Conf. on Computer Vision, pages 894–899.Google Scholar
  45. Rasmussen, C.-E. and Williams, C.K.I. 2006. Gaussian Processes for Machine Learning, MIT Press.Google Scholar
  46. Rathi, Y., Vaswani, N., Tannenbaum, A., and Yezzi, A. 2005. Particle filtering for geometric active contours and application to tracking deforming objects. In IEEE Int. Conf. on Computer Vision and Pattern Recognition, volume2, pp.2–9.Google Scholar
  47. Riklin-Raviv, T., Kiryati, N. and Sochen, N. 2004. Unlevel sets: Geometry and prior-based segmentation. In T.Pajdla and V.Hlavac, editors, European Conf. on Computer Vision, volume 3024 of Lect. Not. Comp. Sci., pp.50–61, Prague. Springer.Google Scholar
  48. Rosenblatt, F. 1956. Remarks on some nonparametric estimates of a density function. Annals of Mathematical Statistics, 27:832–837.MATHMathSciNetGoogle Scholar
  49. Rousson, M., Brox, T., and Deriche, R. 2003. Active unsupervised texture segmentation on a diffusion based feature space. In Proc.IEEE Conf.on Comp.Vision Patt.Recog., pp. 699–704, Madison, WI.Google Scholar
  50. Rousson, M. and Cremers, D. 2005. Efficient kernel density estimation of shape and intensity priors for level set segmentation. In Intl.Conf.on Medical Image Computing and Comp.Ass.Intervention (MICCAI), 1:757–764.CrossRefGoogle Scholar
  51. Rousson, M. and Paragios, N. 2002. Shape priors for level set representations. In A. Heyden etal., editors, Europ. Conf. on Comp. Vis., volume 2351 of Lect. Not. Comp. Sci., pages 78–92. Springer.Google Scholar
  52. Rousson, M., Paragios, N., and Deriche, R. 2004. Implicit active shape models for 3d segmentation in MRI imaging. In Intl.Conf.on Medical Image Computing and Comp.Ass.Intervention (MICCAI), volume 2217 of Lect. Not. Comp. Sci., Springer, pp. 209–216Google Scholar
  53. Silverman, B.W. 1978. Choosing the window width when estimating a density. Biometrika, 65:1–11.MATHMathSciNetCrossRefGoogle Scholar
  54. Silverman, B.W. 1992. Density estimation for statistics and data analysis. Chapman and Hall, London.Google Scholar
  55. Sussman, M., Smereka P., and Osher S.J. 1994. A level set approach for computing solutions to incompressible twophase flow. J. of Comp. Phys., 94:146–159.CrossRefGoogle Scholar
  56. Trouvé, A. 1998. Diffeomorphisms, groups and pattern matching in image analysis. Int.J.ofComputer Vision, 28(3):213–21.CrossRefGoogle Scholar
  57. Tsai, A., Yezzi, A., Wells, W., Tempany, C., Tucker, D., Fan, A., Grimson, E., and Willsky A. 2001. Model–based curve evolution technique for image segmentation. In Comp.Vision Patt.Recog., Kauai, Hawaii, pp. 463–468.Google Scholar
  58. Tsai, A., Yezzi, A.J., and Willsky, A.S. 2001. Curve evolution implementation of the Mumford-Shah functional for image segmentation, denoising, interpolation, and magnification. IEEE Trans. on Image Processing, 10(8):1169–1186.CrossRefMATHGoogle Scholar
  59. Wagner, T.J. 1975. Nonparametric estimates of probability densities. IEEE Trans. on Inform. Theory, 21:438–440.MATHCrossRefGoogle Scholar
  60. Yezzi, A. and Soatto, S. 2003. Deformotion: Deforming motions and shape averages. Int.J.ofComputer Vision, 53(2):153–167.CrossRefGoogle Scholar
  61. Younes, L. 1998. Computable elastic distances between shapes. SIAM J. Appl. Math., 58(2):565–586.MATHMathSciNetCrossRefGoogle Scholar
  62. Zhao, H.-K., Chan, T., Merriman, B. and Osher, S. 1996. A variational level set approach to multiphase motion. J. of Comp. Phys., 127:179–195.MathSciNetCrossRefMATHGoogle Scholar

Copyright information

© Springer Science + Business Media, LLC 2006

Authors and Affiliations

  • Daniel Cremers
    • 1
  • Stanley J. Osher
    • 2
  • Stefano Soatto
    • 3
  1. 1.Department of Computer ScienceUniversity of BonnGermany
  2. 2.Department of MathematicsUniversity of CaliforniaLos Angeles
  3. 3.Department of Computer ScienceUniversity of CaliforniaLos Angeles

Personalised recommendations