Abstract
Many manifold learning procedures try to embed a given feature data into a flat space of low dimensionality while preserving as much as possible the metric in the natural feature space. The embedding process usually relies on distances between neighboring features, mainly since distances between features that are far apart from each other often provide an unreliable estimation of the true distance on the feature manifold due to its non-convexity. Distortions resulting from using long geodesics indiscriminately lead to a known limitation of the Isomap algorithm when used to map non-convex manifolds. Presented is a framework for nonlinear dimensionality reduction that uses both local and global distances in order to learn the intrinsic geometry of flat manifolds with boundaries. The resulting algorithm filters out potentially problematic distances between distant feature points based on the properties of the geodesics connecting those points and their relative distance to the boundary of the feature manifold, thus avoiding an inherent limitation of the Isomap algorithm. Since the proposed algorithm matches non-local structures, it is robust to strong noise. We show experimental results demonstrating the advantages of the proposed approach over conventional dimensionality reduction techniques, both global and local in nature.
Similar content being viewed by others
References
Aharon, M., & Kimmel, R. (2006). Representation analysis and synthesis of lip images using dimensionality reduction. International Journal of Computer Vision, 67(3), 297–312.
Belkin, M., & Niyogi, P. (2002). Laplacian eigenmaps and spectral techniques for embedding and clustering. In Dietterich, T. G., Becker, S., & Ghahramani, Z. (Eds.), Advances in neural inf. proc. sys. (Vol. 14, pp. 585–591). Cambridge: MIT Press.
Belton, D., & Lichti, D. D. (2007). Classification and segmentation of terrestrial laserscanner point clouds using local variance information. ISPRS Image Engineering and Vision Metrology, 61(5), 307–324.
Ben-Tal, A., & Nemirovski, A. S. (2001). Lectures on modern convex optimization: analysis, algorithms, and engineering applications. Philadelphia: Society for Industrial and Applied Mathematics.
Bernstein, M., de Silva, V., Langford, J. C., & Tenenbaum, J. B. (2001). Graph approximations to geodesics on embedded manifolds (Technical Report). Stanford University.
Blake, A., & Zisserman, A. (1987). Visual reconstruction. Cambridge: MIT Press.
Borg, I., & Groenen, P. (1997). Modern multidimensional scaling: theory and applications. New York: Springer.
Boult, T. E., & Kender, J. R. (1986). Visual surface reconstruction using sparse depth data. In Computer vision and pattern recognition (Vol. 86, pp. 68–76).
Brand, M. (2005). Nonrigid embeddings for dimensionality reduction. In Gama, J., Camacho, R., Brazdil, P., Jorge, A., & Torgo, L. (Eds.), Lecture notes in computer science : Vol. 3720. ECML (pp. 47–59). Berlin: Springer.
Brandes, U., & Pich, C. (2007). Eigensolver methods for progressive multidimensional scaling of large data. In Kaufmann, M., & Wagner, D. (Eds.), Graph drawing, Karlsruhe, Germany, September 18–20, 2006 (pp. 42–53). Berlin: Springer.
Bronstein, A. M., Bronstein, M. M., & Kimmel, R. (2005). Three-dimensional face recognition. International Journal of Computer Vision, 64(1), 5–30.
Bronstein, A. M., Bronstein, M. M., & Kimmel, R. (2006a). Generalized multidimensional scaling: a framework for isometry-invariant partial surface matching. Proceedings of the National Academy of Science of the USA, 103(5), 1168–1172.
Bronstein, M. M., Bronstein, A. M., Kimmel, R., & Yavneh, I. (2006b). Multigrid multidimensional scaling. Numerical Linear Algebra with Applications, 13(2–3), 149–171. Special issue on multigrid methods.
Cabay, S., & Jackson, L. (1976). A polynomial extrapolation method for finding limits and antilimits of vector sequences. SIAM Journal on Numerical Analysis, 13, 734–752.
Chalmers, M. (1996). A linear iteration time layout algorithm for visualising high-dimensional data. In IEEE visualization (pp. 127–132).
Chazal, F., Cohen-Steiner, D., & Mérigot, Q. (2007). Stability of boundary measures (Research Report 6219). INRIA Sophia Antipolis.
Choi, H., & Choi, S. (2007). Robust kernel isomap. Pattern Recognition, 40(3), 853–862.
Coifman, R. R., Lafon, S., Lee, A. B., Maggioni, M., Nadler, B., Warner, F., & Zucker, S. W. (2005). Geometric diffusions as a tool for harmonic analysis and structure definition of data. Proceedings of the National Academy of Science of the USA, 102(21), 7426–7431.
Cormen, T. H., Leiserson, C. E., & Rivest, R. L. (1990). Introduction to algorithms. Cambridge/New York: MIT Press/McGraw-Hill.
de Leeuw, J. (1977). Applications of convex analysis to multidimensional scaling. In Barra, J., Brodeau, F., Romier, G., & Custem, B. V. (Eds.), Recent developments in statistics (pp. 133–146). Amsterdam: North Holland.
de Leeuw, J. (1984). Differentiability of Kruskal’s stress at a local minimum. Psychometrika, 49(1), 111–113.
de Leeuw, J. (1988). Convergence of the majorization method for multidimensional scaling. Journal of Classification, 5(2), 163–180.
de Silva, V., & Tenenbaum, J. B. (2002). Global versus local methods in nonlinear dimensionality reduction. In Becker, S., Thrun, S., & Obermayer, K. (Eds.), Advances in neural inf. proc. sys (pp. 705–712). Cambridge: MIT Press.
de Silva, V., & Tenenbaum, J. B. (2004). Sparse multidimensional scaling using landmark points (Tech. rep.). Stanford University.
Diaz, R. M., & Arencibia, A. Q. (Eds.) (2003). Coloring of DT-MRI fiber traces using Laplacian eigenmaps. Berlin: Springer.
Dijkstra, E. W. (1959). A note on two problems in connection with graphs. Numerische Mathematik, 1, 269–271.
Duda, RO, Hart, PE, & Stork, D. G. (2000). Pattern classification and scene analysis (2nd ed.). New York: Wiley.
Eddy, R. P. (1979). Extrapolationg to the limit of a vector sequence. In Wang, P. C. (Ed.), Information linkage between applied mathematics and industry (pp. 387–396). San Diego: Academic Press.
Elad, A., & Kimmel, R. (2003). On bending invariant signatures for surfaces. IEEE Transactions on Pattern Analysis and Machine Intelligence, 25(10), 1285–1295.
Eldar, Y., Lindenbaum, M., Porat, M., & Zeevi, Y. (1997). The farthest point strategy for progressive image sampling. IEEE Transactions on Image Processing, 6(9), 1305–1315.
Freedman, D. (2002). Efficient simplicial reconstructions of manifolds from their samples. IEEE Transactions on Pattern Analysis and Machine Intelligence, 24(10), 1349–1357.
Gonzalez, T. F. (1985). Clustering to minimize the maximum intercluster distance. Theoretical Computer Science, 38, 293–306.
Gopi, M. (2002). On sampling and reconstructing surfaces with boundaries. In Canadian conference on computational geometry (pp. 49–53).
Grimes, C., & Donoho, D. L. (2002). When does Isomap recover the natural parameterization of families of articulated images? (Tech. Rep. 2002-27). Department of Statistics, Stanford University, Stanford, CA 94305-4065.
Grimes, C., & Donoho, D. L. (2003). Hessian eigenmaps: locally linear embedding techniques for high-dimensional data. Proceedings of the National Academy of Sciences of the USA, 100(10), 5591–5596.
Guttman, L. (1968). A general nonmetric technique for finding the smallest coordinate space for a configuration of points. Psychometrika, 33, 469–506.
Guy, G., & Medioni, G. (1997). Inference of surfaces, 3D curves, and junctions from sparse, noisy, 3D data. IEEE Transactions on Pattern Analysis and Machine Intelligence, 19(11), 1265–1277.
Haque, A., & Dilts, G. A. (2007). Three-dimensional boundary detection for particle methods. Journal of Computational Physics, 226(2), 1710–1730.
Hochbaum, D., & Shmoys, D. (1985). A best possible approximation for the k-center problem. Mathematics of Operations Research, 10(2), 180–184.
Kearsley, A., Tapia, R., & Trosset, M. W. (1998). The solution of the metric stress and sstress problems in multidimensional scaling using Newton’s method. Computational Statistics, 13(3), 369–396.
Keller, Y., Lafon, S., & Krauthammer, M. (2005). Protein cluster analysis via directed diffusion. In The fifth Georgia tech international conference on bioinformatics.
Kimmel, R., & Sethian, JA (1998). Computing geodesic paths on manifolds. Proceedings of the National Academy of Sciences of the USA, 95(15), 8431–8435.
Mes̆ina, M. (1977). Convergence acceleration for the iterative solution of the equations X=AX+f. Computer Methods in Applied Mechanics and Engineering, 10, 165–173.
Mordohai, P., & Medioni, G. (2005). Unsupervised dimensionality estimation and manifold learning in high-dimensional spaces by tensor voting. In Proceedings of international joint conference on artificial intelligence (pp. 798–803).
Platt, J. C. (2004). Fastmap, metricmap, and landmark MDS are all nystrom algorithms (Tech. Rep. MSR-TR-2004-26). Microsoft Research (MSR).
Pless, R. (2003). Using Isomap to explore video sequences. In International conference on computer vision, Nice, France (pp. 1433–1440).
Rosman, G., Bronstein, A. M., Bronstein, M. M., & Kimmel, R. (2006). Topologically constrained isometric embedding. In Proc. conf. on machine learning and pattern recognition (MLPR).
Rosman, G., Bronstein, A. M., Bronstein, M. M., Sidi, A., & Kimmel, R. (2008). Fast multidimensional scaling using vector extrapolation (Technical Report CIS-2008-01). Technnion, Israel Institute of Technology.
Rosman, G., Bronstein, A. M., Bronstein, M. M., & Kimmel, R. (2009). Nonlinear dimensionality reduction by topologically constrained isometric embedding (Technical Report CIS-2009-05). Technion.
Roweis, S. T., & Saul, L. K. (2000). Nonlinear dimensionality reduction by locally linear embedding. Science, 290, 2323–2326.
Schwartz, EL, Shaw, A., & Wolfson, E. (1989). A numerical solution to the generalized mapmaker’s problem: flattening nonconvex polyhedral surfaces. IEEE Transactions on Pattern Analysis and Machine Intelligence, 11, 1005–1008.
Sidi, A. (1991). Efficient implementation of minimal polynomial and reduced rank extrapolation methods. Journal of Computational and Applied Mathematics, 36(3), 305–337.
Smith, D. A., Ford, W. F., & Sidi, A. (1987). Extrapolation methods for vector sequences. SIAM Review, 29(2), 199–233.
Tenenbaum, J. B., de Silva, V., & Langford, J. C. (2000). A global geometric framework for nonlinear dimensionality reduction. Science, 290(5500), 2319–2323.
Tong, W. S., Tang, C. K., Mordohai, P., & Medioni, G. (2004). First order augmentation to tensor voting for boundary inference and multiscale analysis in 3d. IEEE Transactions on Pattern Analysis and Machine Intelligence, 26(5), 594–611.
Trosset, M., & Mathar, R. (1997). On the existence of nonglobal minimizers of the stress criterion for metric multidimensional scaling. In American statistical association: proceedings statistical computing section (pp. 158–162).
Weinberger, K. Q., & Saul, L. K. (2004). Unsupervised learning of image manifolds by semidefinite programming. In Computer vision and pattern recognition, Washington, DC (Vol. 2, pp. 988–995).
Weinberger, K. Q., Packer, B. D., & Saul, L. K. (2005). Nonlinear dimensionality reduction by semidefinite programming and kernel matrix factorization. In: Proceedings of the 10th international workshop on artificial intelligence and statistics, Barbados.
Williams, C. K. I. (2002). On a connection between kernel PCA and metric multidimensional scaling. Machine Learning, 46(1–3), 11–19.
Zigelman, G., Kimmel, R., & Kiryati, N. (2002). Texture mapping using surface flattening via multidimensional scaling. IEEE Transactions on Visualization and Computer Graphics, 8(2), 198–207.
Author information
Authors and Affiliations
Corresponding author
Additional information
This research was partly supported by United States–Israel Binational Science Foundation grant No. 2004274, by the Israel Science Foundation (grant no. 623/08) by the Ministry of Science grant No. 3-3414, by the Office of Naval Research (ONR) grant, and by the Elias Fund for Medical Research.
Rights and permissions
About this article
Cite this article
Rosman, G., Bronstein, M.M., Bronstein, A.M. et al. Nonlinear Dimensionality Reduction by Topologically Constrained Isometric Embedding. Int J Comput Vis 89, 56–68 (2010). https://doi.org/10.1007/s11263-010-0322-1
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s11263-010-0322-1