Abstract
Let \(\mathcal{M}\) be an m-dimensional smooth compact manifold embedded in \(\mathbb {R}^d\), where m is a constant known to us. Suppose that a dense set of points are sampled from \(\mathcal{M}\) according to a Poisson process with an unknown parameter. Let p be any sample point, let \(\varrho \) be the local feature size at p, and let \(\varrho \varepsilon \) be the distance from p to the \((n+1)\)th nearest sample point for some n between \(\left( {\begin{array}{c}m+1\\ 2\end{array}}\right) + 1\) and \(\left( {\begin{array}{c}d+1\\ 2\end{array}}\right) \). Using the n sample points nearest to p, we can estimate the tangent space at p and it holds with probability \(1 - O(n^{-1/3})\) that the angular error is \(O(\varepsilon ^2)\). The running time is bounded by the time to compute the thin SVD of an \(n \times \left( {\begin{array}{c}d+1\\ 2\end{array}}\right) \) matrix and the full SVD of an \(n \times d\) matrix, which is usually \(O(d^2n^2)\) in practice. We implemented the algorithm and experimentally verified its effectiveness on both noiseless and noisy data.
Similar content being viewed by others
Notes
Since our method reduces to solving an eigenvalue problem, an appropriate thresholding of the eigenvalues should determine m. We do not pursue automatic dimension detection in this article in order to focus on the tangent estimation. We comment on the determination of m further in the conclusion.
References
Belkin, M., Niyogi, P.: Laplacian eigenmaps and spectral techniques for embedding and clustering. Adv. Neural Inf. Process. Syst. 14, 585–591 (2002)
Belkin, M., Sun, J., Wang, Y.: Constructing Laplace operator from point clouds in \(R^d\). In: Proceedings of the 20th Annual ACM–SIAM Symposium on Discrete Algorithms, pp. 1021–1040 (2009)
Boissonnat, J.-D., Ghosh, A.: Manifold reconstruction using tangential Delaunay complexes. In: Proceedings of the 26th Annual Symposium on Computational Geometry, pp. 324–333 (2010)
Boissonnat, J.-D., Guibas, L.J., Oudot, S.Y.: Manifold reconstruction in arbitrary dimensions using witness complexes. Discrete Comput. Geom. 42, 37–70 (2009)
Boxma, O.J., Yechiali, U.: Poisson Processes, ordinary and compound. In: Ruggeri, F., Kenett, R.S., Faltin, F.W. (eds.) Encyclopedia of Statistics in Quality and Reliability. Wiley, New York (2007)
Carter, K.M., Raich, R., Hero, A.O.: On local intrinsic dimension estimation and its applications. IEEE Trans. Signal Process. 58, 650–663 (2010)
Chan, T.F.: An improved algorithm for computing the singular value decomposition. ACM Trans. Mathe. Softw. 8, 72–83 (1982)
Cheng, S.-W., Chiu, M-K.: Dimension detection via slivers. In: Proceedings of the 20th Annual ACM–SIAM Symposium on Discrete Algorithms, pp. 1001–1010 (2009)
Cheng, S.-W., Dey, T.K., Ramos, E.A.: Manifold reconstruction from point samples. In: Proceedings of the 16th Annual ACM–SIAM Symposium on Discrete Algorithms, pp. 1018–1027 (2005)
Cheng, S.-W., Wang, Y., Wu, Z.: Provable dimension detection using principal component analysis. Int. J. Comput. Geom. Appl. 18, 414–440 (2008)
Dey, T.K., Giesen, J., Goswami, S., Zhao, W.: Shape dimension and approximation from samples. Discrete Comput. Geom. 29, 419–434 (2003)
Eisenstat, S.C., Ipsen, I.C.F.: Relative perturbation bounds for eigenspaces and singular vector subspaces. In: Proceedings of the 5th SIAM Conference on Applied Linear Algebra, pp. 62–66 (1994)
Feingold, D.G., Varga, R.S.: Block diagonally dominant matrices and generalizations of the Gershgorin circle theorem. Pac. J. Math. 12, 1241–1250 (1962)
Gashler, M., Martinez, T.: Tangent space guided intelligent neighbor finding. In: Proceedings of International Joint Conference on Neural Networks, pp. 2617–2624 (2011)
Giesen, J., Wagner, U.: Shape dimension and intrinsic metric from samples of manifolds with high codimension. Discrete Comput. Geom. 32, 245–267 (2004)
Golub, G.H., Reinsch, C.: Singular value decomposition and least square solutions. In: Wilkinson, J.H., Reinsch, C. (eds.) Handbook for Automatic Computation, II, Linear Algebra. Springer, New York (1971)
Golub, G.H., van Loan, C.F.: Matrix Computations. Johns Hopkins University Press, Baltimore (1996)
Gong, D., Zhao, X., Medioni, G.: Robust multiple manifolds structure learning. In: Proceedings of the International Conference on Machine Learning (2012)
Gradshtein, I.S., Ryzhik, I.M.: Table of Integrals, Series, and Products. Academic Press, New York (1994)
Hein, M., Audibert, J.-Y.: Intrinsic dimensionality estimation of submanifolds in Euclidean space. In: Proceedings of the 22nd International Conference on Machine Learning, 289–296 (2005)
Kégl, B.: Intrinsic dimension estimation using packing numbers. Adv. Neural Inf. Process. Syst. 14, 681–688 (2003)
Lang, S.: Calculus of Several Variables. Springer, New York (1987)
Le Gall, F.: Faster algorithms for rectangular matrix multiplication. In: Proceedings of the 53rd IEEE Annual Symposium on Foundations of Computer Science, pp. 514–523 (2012)
Levina, E., Bickel, P.J.: Maximum likelihood estimation of intrinsic dimension. Adv. Neural Inf. Process. Syst. 17, 777–784 (2005)
Little, A.V., Maggioni, M., Rosasco, L.: Multiscale geometric methods for data sets I: multiscale SVD, noise and curvature. Computer Science and Artificial Intelligence Laboratory Technical Report, MIT-CSAIL-TR-2012-029, CBCL-310, 8 Sept 2012
massoud Farahmand, A., Szepesvári, C., Audibert, J.-Y.: Manifold-adaptive dimension estimation. In: Proceedings of the 24th International Conference on Machine Learning, pp. 265–272 (2007)
Morgan, F.: Riemannian Geometry: A Beginner’s Guide. A. K. Peters, Wellesley, MA (1998)
Nascimento, J.C., Silva, J.G.: Manifold learning for object tracking with multiple motion dynamics. In: Proceedings of ECCV 2010, Part III. LNCS, vol. 6313 pp. 172–185 (2010)
Roweis, S., Saul, L.: Nonlinear dimensionality reduction by locally linear embedding. Science 290, 2323–2326 (2000)
Schölkopf, B., Smola, A., Müller, K.-R.: Nonlinear component analysis as a kernel eigenvalue problem. Neural Comput. 10, 1299–1319 (1998)
Taubin, G.: Estimation of planar curves, surfaces and nonplanar space curves defined by implicit equations, with applications to edge and range image segmentation. IEEE Trans. Pattern Anal. Mach. Intell. 13, 1115–1138 (1991)
Tenenbaum, J.B., de Silva, V., Langford, J.C.: A global geometric framework for nonlinear dimensionality reduction. Science 290, 2319–2323 (2000)
Acknowledgments
We thank the anonymous referees for their helpful comments and useful advice. Research supported by the Research Grant Council, Hong Kong, China (Project No. 612109). Part of the work was done while Chiu was at HKUST. Man-Kwun Chiu: JST, ERATO, Kawarabayashi Large Graph Project.
Author information
Authors and Affiliations
Corresponding author
Additional information
Editor in Charge: Kenneth Clarkson
Rights and permissions
About this article
Cite this article
Cheng, SW., Chiu, MK. Tangent Estimation from Point Samples. Discrete Comput Geom 56, 505–557 (2016). https://doi.org/10.1007/s00454-016-9809-z
Received:
Revised:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s00454-016-9809-z