Balanced Exploration and Exploitation Model Search for Efficient Epipolar Geometry Estimation

  • Liran Goshen
  • Ilan Shimshoni
Part of the Lecture Notes in Computer Science book series (LNCS, volume 3952)


The estimation of the epipolar geometry is especially difficult where the putative correspondences include a low percentage of inlier correspondences and/or a large subset of the inliers is consistent with a degenerate configuration of the epipolar geometry that is totally incorrect. This work presents the Balanced Exploration and Exploitation Model Search (BEEM) algorithm that works very well especially for these difficult scenes.

The BEEM algorithm handles the above two difficult cases in a unified manner. The algorithm includes the following main features: (1) Balanced use of three search techniques: global random exploration, local exploration near the current best solution and local exploitation to improve the quality of the model. (2) Exploits available prior information to accelerate the search process. (3) Uses the best found model to guide the search process, escape from degenerate models and to define an efficient stopping criterion. (4) Presents a simple and efficient method to estimate the epipolar geometry from two SIFT correspondences. (5) Uses the locality-sensitive hashing (LSH) approximate nearest neighbor algorithm for fast putative correspondences generation.

The resulting algorithm when tested on real images with or without degenerate configurations gives quality estimations and achieves significant speedups compared to the state of the art algorithms!


Fundamental Matrix Singularity Constraint Inlier Rate Point Correspondence Sift Descriptor 
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.
    Fischler, M., Bolles, R.: Random sample consensus: A paradigm for model fitting with applications to image analysis and automated cartography. Comm. of the ACM 24, 381–395 (1981)MathSciNetCrossRefGoogle Scholar
  2. 2.
    Torr, P.: Motion segmentation and outlier detection. PhD thesis, Dept. of Engineering Science, University of Oxford (1995)Google Scholar
  3. 3.
    Tordoff, B., Murray, D.W.: Guided sampling and consensus for motion estimation. In: Heyden, A., Sparr, G., Nielsen, M., Johansen, P. (eds.) ECCV 2002. LNCS, vol. 2350, pp. 82–96. Springer, Heidelberg (2002)CrossRefGoogle Scholar
  4. 4.
    Chum, O., Matas, J.: Matching with PROSAC: Progressive sample consensus. In: CVPR, pp. I:220–226 (2005)Google Scholar
  5. 5.
    Chum, O., Matas, J., Kittler, J.: Locally optimized RANSAC. In: German Pattern Recognition Symposium, pp. 236–243 (2003)Google Scholar
  6. 6.
    Goshen, L., Shimshoni, I.: Guided sampling via weak motion models and outlier sample generation for epipolar geometry estimation. In: CVPR, pp. I:1105–1112 (2005)Google Scholar
  7. 7.
    Schaffalitzky, F., Zisserman, A.: Multi-view matching for unordered image sets, or how do I organize my holiday snaps? In: Heyden, A., Sparr, G., Nielsen, M., Johansen, P. (eds.) ECCV 2002. LNCS, vol. 2350, pp. 414–431. Springer, Heidelberg (2002)CrossRefGoogle Scholar
  8. 8.
    Chum, O., Matas, J., Obdrzalek, S.: Enhancing RANSAC by generalized model optimization. In: ACCV, pp. II:812–817 (2004)Google Scholar
  9. 9.
    Chum, O., Werner, T., Matas, J.: Two-view geometry estimation unaffected by a dominant plane. In: CVPR, pp. I:772–779 (2005)Google Scholar
  10. 10.
    Gionis, A., Indyk, P., Motwai, R.: Similarity search in high dimensions via hashing. In: ICVL, pp. 518–529 (1999)Google Scholar
  11. 11.
    Grabner, M., Grabner, H., Bischof, H.: Fast approximated sift. In: Asian Conference on Computer Vision, pp. 918–927 (2006)Google Scholar
  12. 12.
    Chen, H., Meer, P.: Robust regression with projection based m-estimators. In: International Conference on Computer Vision, pp. 878–885 (2003)Google Scholar
  13. 13.
    Rozenfeld, S., Shimshoni, I.: The modified pbM-estimator method and a runtime analysis technique for the ransac family. In: CVPR, pp. I:1113–1120 (2005)Google Scholar
  14. 14.
    Lowe, D.: Distinctive image features from scale-invariant keypoints. IJCV 60, 91–110 (2004)CrossRefGoogle Scholar
  15. 15.
    Hartley, R.: In defense of the eight-point algorithm. IEEE Trans. Patt. Anal. Mach. Intell. 19, 580–593 (1997)CrossRefGoogle Scholar
  16. 16.
    Georgescu, B., Shimshoni, I., Meer, P.: Mean shift based clustering in high dimensions: A texture classification example. In: ICCV, pp. 456–463 (2003)Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2006

Authors and Affiliations

  • Liran Goshen
    • 1
  • Ilan Shimshoni
    • 2
  1. 1.Faculty of Industrial Engineering & ManagementTechnionHaifaIsrael
  2. 2.Department of Management Information SystemsHaifa UniversityHaifaIsrael

Personalised recommendations