Uncertainty Analysis Using Geometrical Property Between 2D-to-3D Under Affine Projection

  • Sungshik Koh
  • Phil Jung Kim
Conference paper
Part of the Lecture Notes in Computer Science book series (LNCS, volume 4223)


In this paper, we propose uncertainty analysis using geometrical property between 2D-to-3D under affine reconstruction. In situations when are no missing data in an observation matrix, the accurate solution is known to be provided by Singular Value Decomposition (SVD). However, when converting image sequences to 3D, several entries of the matrix have not been observed and other entries have been perturbed by the influence of noise. In this case, there is no simple solution. In this paper, a new approach is applied for recovering missing data using geometrical properties between a 2D image plane and 3D shape and for estimating noise level in an observation matrix using ranks of SVD. This paper consists of four main phases: geometrical properties between 2D image plane and 3D error space, geometrical recovering of missing data, and noise level estimation in the observation matrix.


Noise Level Geometrical Property Singular Value Decomposition Factorization Method Camera Motion 
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.


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.


  1. 1.
    Tomasi, Kanade, T.: Shape and motion from image streams under orthography: A factorization method. Int. J. of Computer Vision 9(2), 137–154 (1992)CrossRefGoogle Scholar
  2. 2.
    Sturm, P., Triggs, B.: A factorization based algorithm for multi-image projective structure and motion. In: Buxton, B.F., Cipolla, R. (eds.) ECCV 1996. LNCS, vol. 1065, pp. 709–720. Springer, Heidelberg (1996)Google Scholar
  3. 3.
    Poelman, C.J., Kanade, T.: A paraperspective factorization method for shape and motion recovery. IEEE Trans. on PAMI 19(3), 206–218 (1997)Google Scholar
  4. 4.
    Morita, T., Kanade, T.: A sequential factorization method for recovering shape and motion from image streams. IEEE Trans. on PAMI 19(8), 858–867 (1997)Google Scholar
  5. 5.
    Aguiar, P.M.Q., Moura, J.M.F.: Rank 1 Weighted Factorization for 3D Structure Recovery: Algorithms and Performance Analysis. IEEE Trans. on PAMI 25(9), 1134–1149 (2003)Google Scholar
  6. 6.
    Irani, M., Anandan, P.: Factorization with uncertainty. In: Vernon, D. (ed.) ECCV 2000. LNCS, vol. 1842, pp. 539–553. Springer, Heidelberg (2000)CrossRefGoogle Scholar
  7. 7.
    Shapiro, L.: Affine Analysis of Image Sequences. Cambridge University Press, Cambridge (1995)MATHCrossRefGoogle Scholar
  8. 8.
    Quan, L., Kanade, T.: A factorization method for affine structure from line correspondences. In: Proc. IEEE CVPR, San Francisco CA, USA, pp. 803–808 (1996)Google Scholar
  9. 9.
    Sun, Z., Ramesh, V., Murat Tekalp, A.: Error characterization of the factorization method. Computer Vision and Image Understanding 82, 110–137 (2001)MATHCrossRefGoogle Scholar
  10. 10.
    Daniilidis, K., Spetsakis, M.: Understanding noise sensitivity in structure from motion. Visual Navigation, 61–88 (1996)Google Scholar
  11. 11.
    Morris, D.D., Kanatani, K., Kanade, T.: Uncertainty modeling for optimal structure from motion. In: Triggs, B., Zisserman, A., Szeliski, R. (eds.) ICCV-WS 1999. LNCS, vol. 1883, pp. 200–217. Springer, Heidelberg (2000)CrossRefGoogle Scholar
  12. 12.
    Szeliski, R., Kang, S.B.: Shape ambiguities in structure-from-motion. IEEE Trans. on PAMI 19, 506–512 (1997)Google Scholar
  13. 13.
    Szeliski, R., Kang, S.B.: Shape Ambiguities in Structure From Motion. Technical Report 96/1, Digital Equipment Corporation, Cambridge Research Lab, Cambridge, Mass (1996)Google Scholar
  14. 14.
    Harris, C.J., Stephens, M.: A Combined Corner and Edge Detector. In: Proc. Alvey Vision Conf., pp. 147–151 (1998)Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2006

Authors and Affiliations

  • Sungshik Koh
    • 1
  • Phil Jung Kim
    • 2
  1. 1.Insan Innovation Telecom Co., Ltd.Korea
  2. 2.Dept. of ITSunghwa CollegeKorea

Personalised recommendations