Estimating the Jacobian of the Singular Value Decomposition: Theory and Applications

  • Théodore Papadopoulo
  • Manolis I. A. Lourakis
Conference paper
Part of the Lecture Notes in Computer Science book series (LNCS, volume 1842)


The Singular Value Decomposition (SVD) of a matrix is a linear algebra tool that has been successfully applied to a wide variety of domains. The present paper is concerned with the problem of estimating the Jacobian of the SVD components of a matrix with respect to the matrix itself. An exact analytic technique is developed that facilitates the estimation of the Jacobian using calculations based on simple linear algebra. Knowledge of the Jacobian of the SVD is very useful in certain applications involving multivariate regression or the computation of the uncertainty related to estimates obtained through the SVD. The usefulness and generality of the proposed technique is demonstrated by applying it to the estimation of the uncertainty for three different vision problems, namely self-calibration, epipole computation and rigid motion estimation.


Singular Value Decomposition Machine Intelligence Image Watermark Fundamental Matrix Intrinsic Parameter 
These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.


  1. 1.
    E. Anderson, Z. Bai, C. Bishof, J. Demmel, J. Dongarra, J. D. Croz, A. Greenbaum, S. Hammarling, A. McKenney, S. Ostrouchov, and D. Sorensen. LAPACK Users’ Guide. Society for Industrial and Applied Mathematics, 3600 University City Science Center, Philadelphia, PA 19104-2688, second edition, 1994.Google Scholar
  2. 2.
    K. Arun, T. Huang, and S. Blostein. Least-squares fitting of two 3-D point sets. IEEE Transactions on Pattern Analysis and Machine Intelligence, 9(5):698–700, Sept. 1987.Google Scholar
  3. 3.
    Å. Björck. Numerical methods for least squares problems.SIAM, 1996.Google Scholar
  4. 4.
    T. Chen, S. Amari, and Q. Lin. A unified algorithm for principal and minor components extraction. Neural Networks, 11(3):385–390, 1998.CrossRefGoogle Scholar
  5. 5.
    G. Csurka, C. Zeller, Z. Zhang, and O. Faugeras. Characterizing the uncertainty of the fundamental matrix. CVGIP: Image Understanding, 68(1):18–36, Oct. 1997.Google Scholar
  6. 6.
    J. Demmel and K. Veselić. Jacobi’s method is more accurate than QR. SIAM Journal on Matrix Analysis and Applications, 13:1204–1245, 1992.zbMATHCrossRefMathSciNetGoogle Scholar
  7. 7.
    K. Diamantaras and S. Kung. Multilayer neural networks for reduced-rank approximation. IEEE Trans. on Neural Networks, 5(5):684–697, 1994.CrossRefGoogle Scholar
  8. 8.
    O. Faugeras. Three-Dimensional Computer Vision: a Geometric Viewpoint.MIT Press, 1993.Google Scholar
  9. 9.
    N. Georgis, M. Petrou, and J. Kittler. On the correspondence problem for wide angular separation of non-coplanar points. Image and Vision Computing, 16:35–41, 1998.Google Scholar
  10. 10.
    G. Golub and C. V. Loan. Matrix computations. The John Hopkins University Press, Baltimore, Maryland, second edition, 1989.zbMATHGoogle Scholar
  11. 11.
    R. Hartley. Kruppa’s equations derived from the fundamental matrix. IEEE Transactions on Pattern Analysis and Machine Intelligence, 19(2):133–135, Feb. 1997.Google Scholar
  12. 12.
    R. I. Hartley. Estimation of relative camera positions for uncalibrated cameras. In G. Sandini, editor, Proceedings of the 2nd European Conference on Computer Vision, pages 579–587, Santa Margherita, Italy, May 1992. Springer-Verlag.Google Scholar
  13. 13.
    R. I. Hartley. Cheirality invariants. In Proceedings of the ARPA Image Understanding Workshop, pages 745–753, Washington, DC, Apr. 1993. Defense Advanced Research Projects Agency, Morgan Kaufmann Publishers, Inc.Google Scholar
  14. 14.
    D. Jones and J. Malik. Computational framework for determining stereo correspondence from a set of linear spatial filters. Image and Vision Computing, 10(10):699–708, 1992.CrossRefGoogle Scholar
  15. 15.
    B. N. K. Konstantinides and G. Yovanof. Noise estimation and filtering using block-based singular value decomposition. IEEE Trans. on Image Processing, 6(3):479-483, 1997.Google Scholar
  16. 16.
    K. Kanatani. Analysis of 3-D rotation fitting. IEEE Transactions on Pattern Analysis and Machine Intelligence, 16(5):543–54, 1994.CrossRefGoogle Scholar
  17. 17.
    K. Kanatani. Statistical Optimization for Geometric Computation: Theory and Practice. Elsevier Science, 1996.Google Scholar
  18. 18.
    R. Liu and T. Tan. A new SVD based image watermarking method. In Proc. of the 4th Asian Conference on Computer Vision, volume I, pages 63–67, Jan. 2000.Google Scholar
  19. 19.
    H. Longuet-Higgins. A computer algorithm for reconstructing a scene from two projections. Nature, 293:133–135, 1981.CrossRefGoogle Scholar
  20. 20.
    M. I. Lourakis and R. Deriche. Camera self-calibration using the singular value decomposition of the fundamental matrix: From point correspondences to 3D measurements. Research Report 3748, INRIA Sophia-Antipolis, Aug. 1999.Google Scholar
  21. 21.
    M. I. Lourakis and R. Deriche. Camera self-calibration using the Kruppa equations and the SVD of the fundamental matrix: The case of varying intrinsic parameters. Research report, INRIA Sophia-Antipolis, 2000. In preparation.Google Scholar
  22. 22.
    M. I. Lourakis and R. Deriche. Camera self-calibration using the singular value decomposition of the fundamental matrix. In Proc. of the 4th Asian Conference on Computer Vision, volume I, pages 403–408, Jan. 2000.Google Scholar
  23. 23.
    Q.-T. Luong and O. Faugeras. Self-calibration of a moving camera from point correspondences and fundamental matrices. The International Journal of Computer Vision, 22(3):261–289, 1997.CrossRefGoogle Scholar
  24. 24.
    Q.-T. Luong and O. Faugeras. On the determination of epipoles using cross-ratios. CVGIP: Image Understanding, 71(1):1–18, July 1998.Google Scholar
  25. 25.
    S. J. Maybank and O. D. Faugeras. A theory of self-calibration of a moving camera. The International Journal of Computer Vision, 8(2):123–152, Aug. 1992.Google Scholar
  26. 26.
    M. Moonen and B. D. Moor, editors. SVD and Signal Processing III: Algorithms, Analysis and Applications. Elsevier, Amsterdam, 1995.Google Scholar
  27. 27.
    T. Morita and T. Kanade. A sequential factorization method for recovering shape and motion from image streams. IEEE Transactions on Pattern Analysis and Machine Intelligence, 19(8):858–867, 1997.CrossRefGoogle Scholar
  28. 28.
    C. Poelman and T. Kanade. A paraperspective factorization method for shape and motion recovery. IEEE Transactions on Pattern Analysis and Machine Intelligence, 19(3):206–218, 1997.CrossRefGoogle Scholar
  29. 29.
    W. H. Press, B. P. Flannery, S. A. Teukolsky, and W. T. Vetterling. Numerical Recipes in C Cambridge University Press, 1988.Google Scholar
  30. 30.
    S. Roy, J. Meunier, and I. Cox. Cylindrical rectification to minimize epipolar distortion. In Proceedings of the International Conference on Computer Vision and Pattern Recognition, pages 393–399, San Juan, Puerto Rico, June 1997. IEEE Computer Society.Google Scholar
  31. 31.
    L. Scharf. The SVD and reduced rank signal processing. Signal Processing, 25(2):113–133, 1991.zbMATHCrossRefGoogle Scholar
  32. 32.
    A. Shashua and N. Navab. Relative affine structure-canonical model for 3D from 2D geometry and applications. IEEE Transactions on Pattern Analysis and Machine Intelligence, 18:873–883, 1996.CrossRefGoogle Scholar
  33. 33.
    G. W. Stewart. Error and perturbation bounds for subspaces associated with certain eigenvalue problems. SIAM Review, 15(4):727–764, Oct. 1973.Google Scholar
  34. 34.
    P. Sturm and B. Triggs. A factorization based algorithm for multi-image projective structure and motion. In B. Buxton, editor, Proceedings of the 4th European Conference on Computer Vision, pages 709–720, Cambridge, UK, Apr. 1996.Google Scholar
  35. 35.
    C. Tomasi and T. Kanade. Shape and motion from image streams under orthography: a factorization method. The International Journal of Computer Vision, 9(2):137–154, 1992.CrossRefGoogle Scholar
  36. 36.
    R. Tsai, T. Huang, and W. Zhu. Estimating three-dimensional motion parameters of a rigid planar patch, II: singular value decomposition. IEEE Transactions on Acoustic, Speech and Signal Processing, 30(4):525–534, 1982.CrossRefGoogle Scholar
  37. 37.
    S. Umeyama. Least-squares estimation of transformation parameters between two point patterns. IEEE Transactions on Pattern Analysis and Machine Intelligence, 13(4):376–380, 1991.CrossRefGoogle Scholar
  38. 38.
    R. J. Vaccaro. A second-order perturbation expansion for the svd. SIAM Journal on Matrix Analysis and Applications, 15(2):661–671, Apr. 1994.Google Scholar
  39. 39.
    C. Wu, M. Berry, S. Shivakumar, and J. McLarty. Neural networks for full-scale protein sequence classification: Sequence encoding with singular value decomposition. Machine Learning, 21(1-2), 1995.Google Scholar
  40. 40.
    J. Yang and C. Lu. Combined techniques of singular value decomposition and vector quantization for image coding. IEEE Trans. on Image Processing, 4(8):1141–1146, 1995.CrossRefGoogle Scholar
  41. 41.
    C. Zeller and O. Faugeras. Camera self-calibration from video sequences: the Kruppa equations revisited. Research Report 2793, INRIA, Feb. 1996.Google Scholar
  42. 42.
    M. Zhang, K. Yu, and R. Haralick. Fast correlation registration method using singular value decomposition. International Journal of Intelligent Systems, 1:181–194, 1986.CrossRefGoogle Scholar
  43. 43.
    Z. Zhang, R. Deriche, O. Faugeras, and Q.-T. Luong. A robust technique for matching two uncalibrated images through the recovery of the unknown epipolar geometry. Artificial Intelligence Journal, 78:87–119, Oct. 1995.Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2000

Authors and Affiliations

  • Théodore Papadopoulo
    • 1
  • Manolis I. A. Lourakis
    • 1
  1. 1.INRIA Sophia AntipolisSophia-antipolisFrance

Personalised recommendations