Feature Matching and Pose Estimation Using Newton Iteration

  • Hongdong Li
  • Richard Hartley
Conference paper
Part of the Lecture Notes in Computer Science book series (LNCS, volume 3617)

Abstract

Feature matching and pose estimation are two crucial tasks in computer vision. The widely adopted scheme is first find the correct matches then estimate the transformation parameters. Unfortunately, such simple scheme does not work well sometimes, because these two tasks of matching and estimation are mutually interlocked. This paper proposes a new method that is able to estimate the transformation and find the correct matches simultaneously. The above interlock is disentangled by an alternating Newton iteration method. We formulate the problem as a nearest-matrix problem, and provide a different numerical technique. Experiments on both synthetic and real images gave good results. Fast global convergence was obtained without the need of good initial guess.

Keywords

Feature Point Real Image Feature Match Newton Iteration Permutation Matrix 
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.

References

  1. 1.
    David, P., Dementhon, D., Duraiswami, R., Samet, H.: SoftPOSIT: Simultaneous pose and correspondence determination. In: Heyden, A., Sparr, G., Nielsen, M., Johansen, P. (eds.) ECCV 2002. LNCS, vol. 2352, pp. 698–714. Springer, Heidelberg (2002)CrossRefGoogle Scholar
  2. 2.
    Gold, S., Pappu, S., Lu, C., Rangarajan, A., Maolsness, E.: New algorithm for 2D and 3D point matching: Pose estimation and correspondence. PR (31), 1019–1031 (1998)Google Scholar
  3. 3.
    Scott, G.L., Longuet-Higgins, H.C.: An Algorithm for Associating the Features of Two Images. Proc. Royal, Soci. London B-244, 1–26 (1991)Google Scholar
  4. 4.
    Shapiro, L.S., Brady, J.M.: Feature-Based Correspondence: An Eigenvector Approach. IVC (10)(5), 283–288 (1992)Google Scholar
  5. 5.
    Scalaroff, S., Pentland, A.: Modal Matching for correspondence and recognition. T-PAMI 17(6), 545–561 (1995)Google Scholar
  6. 6.
    Maciel, Costeira: A global solution to sparse correspondence problems, T-PAMI 25–2 (2003)Google Scholar
  7. 7.
    Kosowsky, J., Yuille, A.: The invisible hand algorithm: solving the assignment problem with statistical physics. Neural networks 7, 477–490 (1994)MATHCrossRefGoogle Scholar
  8. 8.
    Higham, N.: Stable iteration for the matrix square root. Numerical algorithm 15, 227–242 (1997)MATHCrossRefMathSciNetGoogle Scholar
  9. 9.
    Carcassoni, M., Hancock, E.: Spectral correspondence for point pattern matching. PR(36) (1), 193–204 (2003)Google Scholar
  10. 10.
    Rangarajan, A., Chui, H., Bookstein, F.L.: The Softassign Procrustes Matching Algorithm. In: Duncan, J., Gindi, G. (eds.) Information Processing in Medical (1997)Google Scholar
  11. 11.
    Kendall, D.: Shape Manifolds, Procrustean metrics and complex projective spaces. Bullet. London. Math.Society 16, 81–121 (1984)MATHCrossRefMathSciNetGoogle Scholar
  12. 12.
    Hartley, R., Zisserman, A.: Multiple View Geometry in Computer Vision, 2nd edn. Cambridge University Press, Cambridge (2004)MATHGoogle Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2005

Authors and Affiliations

  • Hongdong Li
    • 1
  • Richard Hartley
    • 1
  1. 1.Research School of Information Sciences and EngineeringThe Australian National University, ASSeT, Canberra Research Labs, National ICT Australia 

Personalised recommendations