HIERARCHICAL ESTIMATION OF IMAGE FEATURES WITH COMPENSATION OF MODEL APPROXIMATION ERRORS

Stefano Casadei

Abstract

To facilitate the optimal estimation of the parameters of a complex image feature, the feature’s model is fragmented into simpler approximating models. By repeating this fragmentation procedure recursively, a hierarchy of feature models is obtained. To ensure that feature parameter values are recovered exactly in the limit of high SNR, an algorithm is proposed to compensate for the model approximation errors between adjacent levels of the hierarchy.

References

  1. Baker, S., Nayar, S., and Murase, H. (1998). Parametric feature detection. IJCV, 27:27-50.
  2. Blaszka, T. and Deriche, R. (1994a). A model based method for characterization and location of curved image features. Technical Report 2451, Inria-Sophia.
  3. Blaszka, T. and Deriche, R. (1994b). Recovering and characterizing image features using an efficient model based approach. Technical Report 2422, INRIA.
  4. Canny, J. (1986). A computational approach to edge detection. PAMI, 8:679-698.
  5. Casadei, S. and Mitter, S. K. (1996). A hierarchical approach to high resolution edge contour reconstruction. In CVPR, pages 149-153.
  6. Casadei, S. and Mitter, S. K. (1998). Hierarchical image segmentation - part i: Detection of regular curves in a vector graph. IJCV, 27(3):71-100.
  7. Casadei, S. and Mitter, S. K. (1999a). Beyond the uniqueness assumption : ambiguity representation and redundancy elimination in the computation of a covering sample of salient contour cycles. Computer Vision and Image Understanding.
  8. Casadei, S. and Mitter, S. K. (1999b). An efficient and provably correct algorithm for the multiscale estimation of image contours by means of polygonal lines. IEEE Trans. Information Theory, 45(3).
  9. Deriche, R. and Blaszka, T. (1993). Recovering and characterizing image features using an efficient model based approach. In CVPR.
  10. Deriche, R. and Giraudon, G. (1990). Accurate corner detection: An analytical study. ICCV, 90:66-70.
  11. Deriche, R. and Giraudon, G. (1993). A computational approach for corner and vertex detection. IJCV, 10(2):101-124.
  12. Köthe, U. (2003). Integrated edge and junction detection with the boundary tensor. In ICCV 7803, page 424.
  13. Marr, D. (1982). Vision. W.H.Freeman & Co.
  14. Nalwa, V. S. and Binford, T. O. (1986). On detecting edges. PAMI, 8:699-714.
  15. Nayar, S., Baker, S., and Murase, H. (1996). Parametric feature detection. In CVPR, pages 471-477.
  16. Perona, P. (1995). Deformable kernels for early vision. PAMI, 17:488-499.
  17. Rohr, K. (1992). Recognizing corners by fitting parametric models. IJCV, 9(3).
  18. Steger, C. (1998). An unbiased detector of curvilinear structures. PAMI, 20(2):113-125.
  19. Würtz, R. P. and Lourens, T. (1997). Corner detection in color images by multiscale combination of endstopped cortical cells. In ICANN 7897: Proceedings of the 7th International Conference on Artificial Neural Networks, pages 901-906, London, UK. SpringerVerlag.
Download


Paper Citation


in Harvard Style

Casadei S. (2006). HIERARCHICAL ESTIMATION OF IMAGE FEATURES WITH COMPENSATION OF MODEL APPROXIMATION ERRORS . In Proceedings of the First International Conference on Computer Vision Theory and Applications - Volume 1: VISAPP, ISBN 972-8865-40-6, pages 381-388. DOI: 10.5220/0001369003810388


in Bibtex Style

@conference{visapp06,
author={Stefano Casadei},
title={HIERARCHICAL ESTIMATION OF IMAGE FEATURES WITH COMPENSATION OF MODEL APPROXIMATION ERRORS},
booktitle={Proceedings of the First International Conference on Computer Vision Theory and Applications - Volume 1: VISAPP,},
year={2006},
pages={381-388},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0001369003810388},
isbn={972-8865-40-6},
}


in EndNote Style

TY - CONF
JO - Proceedings of the First International Conference on Computer Vision Theory and Applications - Volume 1: VISAPP,
TI - HIERARCHICAL ESTIMATION OF IMAGE FEATURES WITH COMPENSATION OF MODEL APPROXIMATION ERRORS
SN - 972-8865-40-6
AU - Casadei S.
PY - 2006
SP - 381
EP - 388
DO - 10.5220/0001369003810388