This Article 
 Bibliographic References 
 Add to: 
Cooperative Robust Estimation Using Layers of Support
May 1995 (vol. 17 no. 5)
pp. 474-487

Abstract—We present an approach to the problem of representing images that contain multiple objects or surfaces. Rather than use an edge-based approach to represent the segmentation of a scene, we propose a multilayer estimation framework which uses support maps to represent the segmentation of the image into homogeneous chunks. This support-based approach can represent objects that are split into disjoint regions, or have surfaces that are transparently interleaved. Our framework is based on an extension of robust estimation methods that provide a theoretical basis for support-based estimation. We use a selection criteria derived from the Minimum Description Length principle to decide how many support maps to use in describing an image. Our method has been applied to a number of different domains, including the decomposition of range images into constituent objects, the segmentation of image sequences into homogeneous higher-order motion fields, and the separation of tracked motion features into distinct rigid-body motions.

[1] E.H. Adelson and P. Anandan,“Ordinal characteristics of transparency,” Proc. AAAI Workshop on Qualitative Vision,Boston, Mass., 1990, pp. 77-81.
[2] A.J. Azarbayejani,B. Horowitz,, and A.P. Pentland,“Recursive estimation of structure and motion using the relative orientation constraint,” Proc. IEEE Conf. on Computer Vision and Pattern Recognition,New York, June 1993.
[3] F. Attneave,“Pragnanz and soap-bubble systems: A theoretical exploration,” J. Beck, ed., Organization and Representation in Perception.Hillsdale, N.J.: Erlbaum, 1992.
[4] M.J. Black and P. Anandan,“The robust estimation of multiple motions: Affine and piecewise-smooth flow fields,” Xerox PARC Technical Report SPL-93-092, Dec. 1993.
[5] M.J. Black and P. Anandan, “Robust Dynamic Motion Estimation Over Time,” Proc. IEEE Conf. Computer Vision and Pattern Recognition, pp. 296–302, June 1991.
[6] M.J. Black and P. Anandan,“A framework for the robust estimation of optical flow,” Proc. Int’l Conf. on Computer Vision, ICCV-93, Berlin, pp. 231-236, May 1993.
[7] A.E. Beaton and J.W. Tukey,“The fitting of power series, meaning polynomials, illustrated on band-spectroscopic data,” Technometrics, vol. 16, pp. 147-185, 1974.
[8] P.J. Besl,J. B. Birch,, and L.T. Watson,“Robust window operators,” Proc. 2nd Int’l Conf. on Computer Vision, pp. 591-600, 1988.
[9] A. Blake and A. Zisserman, Visual Reconstruction. MIT Press, 1987.
[10] T. Darrell,S. Sclaroff,, and P. Pentland,“Segmentation by minimal description,” Proc. 3rd Int’l Conf. on Computer Vision, pp. 112-116, 1990.
[11] T. Darrell and A.P. Pentland,“Robust estimation of a multilayer motion representation,” Proc. IEEE Workshop on Visual Motion, pp. 173-177, 1991.
[12] T. Darrell and E.P. Simoncelli,“Separation of transparent motion into layers using velocity-tuned mechanisms,” Proc. Assn. Research in Vision and Opth. Annual Conference (ARVO’93), available as MIT Media Lab Perceptual Computing Technical Report TR-244.
[13] S. Geman and D. Geman,“Stochastic relaxation, Gibbs distribution, and the Bayesian restoration of images,” in IEEE Trans. on Pattern Analysis and Machine Intelligence, vol. 6, pp. 721-741, 1984.
[14] J.E. Hochberg and E. McAlister,“A quantitative approach to figure goodness,” J. Experimental Psychology, vol. 46, pp. 361-364, 1953.
[15] J. Hopfield and D. Tank,“Neural computation of decisions in optimization problems,” Biological Cybernetics, vol. 52, pp. 141-152, 1985.
[16] B.K.P. Horn,Robot Vision.Cambridge, Mass.: MIT Press, 1986.
[17] B.K.P. Horn, “Relative Orientation,” Int'l J. Computer Vision, vol. 4, pp. 59-78, 1990.
[18] P. Huber,Robust Statistical Procedures. SIAM CBMS-NSF series in Appl. Math., vol. 27, 1977.
[19] M. Husain,S. Treue,, and R. Andersen,“Surface interpolation in three-dimensional structure-from-motion perception,” Neural Computation, vol. 1, pp. 324-333, 1989.
[20] K. Koffka,Principles of Gestalt Psychology.New York: Harcourt, Brace&World, 1935.
[21] Y. Leclerc,“Constructing simple stable descriptions for image partitioning,” Int’l J. of Computer Vision, vol. 3, pp.73-102, 1989.
[22] Y. Leclerc,“Region grouping using the minimum-description-length principle,” Proc. DARPA Image Understanding Workshop, pp. 473-479, 1990.
[23] E.L.J. Leeuwenberg,“A perceptual coding language for visual and auditory patterns,” Am. J. of Psychology, vol. 84, pp. 307-399, 1971.
[24] A. Leonardis,“Recover and select paradigm—a robust approach to estimation of parametric models,” Proc. NSF/ARPA Workshop on Performance vs. Methodology in Computer Vision,Seattle, June 1994.
[25] G. Li,“Robust regression,” D.C. Hoaglin, F. Mosteller, and J.W. Tukey, eds., Exploring Data, Tables, Trends and Shapes.New York: John Wiley&Sons, pp. 281-343, 1985.
[26] J.L. Marroquin,“Random measure fields and the integration of visual information,” IEEE Trans. Sys. Man., Cybernetics, vol. 22, pp. 705-716, 1992.
[27] P. Meer, D. Mintz, and A. Rosenfeld, “Robust Regression Methods for Computer Vision: A Review,” Int'l J. Computer Vision, vol. 6, no. 1, pp. 59-70, 1991.
[28] M. Nitzberg and D. Mumford,“The 2.1-D sketch,” Proc. 3rd Int’l Conf. on Computer Vision, pp. 138-144, 1990.
[29] A. Pentland,“Part segmentation for object recognition,” Neural Computation, vol. 1, pp. 82-91, 1989.
[30] A.P. Pentland,“Automatic extraction of deformable part models,” Int’l J. Computer Vision, vol. 4, pp. 107-126, 1990.
[31] A. Pentland and S. Sclaroff, "Closed-Form Solutions for Physically-Based Shape Modeling and Recognition," IEEE Trans. Pattern Analysis and Machine Intelligence, vol. 13, no. 7, pp. 715-729, July 1991.
[32] T. Poggio, V. Torre, and C. Koch, “Computational Vision and Regularization Theory,” Nature, vol. 317, pp. 314-319, 1985.
[33] M. Irani and S. Peleg,“Image sequence enhancement using multiple motions analysis,” Technical Report 91-15, Department of Computer Science Technical Report,The Hebrew University of Jerusalem, Israel, Dec. 1991.
[34] M. Irani, B. Rousso, and S. Peleg, “Computing Occluding and Transparent Motions,” Int'l J. Computer Vision, vol. 12, no. 1, pp. 5-16, Jan. 1994.
[35] W.H. Press, B.P. Flannery, S.A. Teukolsky, and W.T. Vetterling, Numerical Recipes in C.Cambridge, England: Cambridge Univ. Press, 1988.
[36] C.E. Shannon,The Mathematical Theory of Communication.Urbana: Univ. of Illinois Press, 1949.
[37] D. Terzopoulos, "The Computation of Visible Surface Representations," IEEE Trans. Pattern Analysis and Machine Intelligence, vol. 10, no. 4, pp. 417-438, Apr. 1988.
[38] J. Rissanen,“A universal prior for integers and estimation by minimum description length,” The Annals of Statistics, vol. 11, no. 2, pp. 416-431, 1983.
[39] J. Rissanen, Stochastic Complexity in Statistical Inquiry. World Scientific Series in Computer Science, vol. 15, 1989.
[40] E. Simoncelli,E.H. Adelson,, and D.J. Heeger,“Probability distributions of optical flow,” Proc. IEEE Conf. Computer Vision and Pattern Recognition, 1991.
[41] J. Wang and E.H. Adelson,“Layered representations for image sequence coding,” Proc. IEEE Conf. Computer Vision and Pattern Recognition, 1993.
[42] C.S. Wallace and D.M. Boulton,“An information measure for classification,” Computing J., vol. 11, no. 2, pp. 185-195, 1968.

Index Terms:
Segmentation, transparency, robust estimation, perceptual organization, multiple models, range segmentation, motion segmentation, structure from motion.
Trevor Darrell, Alex P. Pentland, "Cooperative Robust Estimation Using Layers of Support," IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 17, no. 5, pp. 474-487, May 1995, doi:10.1109/34.391395
Usage of this product signifies your acceptance of the Terms of Use.