Supervised Learning Based Stereo Matching Using Neural Tree

  • Sanjeev Kumar
  • Asha Rani
  • Christian Micheloni
  • Gian Luca Foresti
Part of the Lecture Notes in Computer Science book series (LNCS, volume 6979)

Abstract

In this paper, a supervised learning based approach is presented to classify tentative matches as inliers or outliers obtained from a pair of stereo images. A balanced neural tree (BNT) is adopted to perform the classification task. A set of tentative matches is obtained using speedup robust feature (SURF) matching and then feature vectors are extracted for all matches to classify them either as inliers or outliers. The BNT is trained using a set of tentative matches having ground-truth information, and then it is used for classifying other sets of tentative matches obtained from the different pairs of images. Several experiments have been performed to evaluate the performance of the proposed method.

Keywords

Neural Tree Stereo Vision Tentative Matches Supervised Learning 

References

  1. 1.
    Mikolajczyk, K., Schmid, C.: A performance evaluation of local descriptors. IEEE Trans. Pattern Anal. Mach. Intell. 27(10), 1615–1630 (2005)CrossRefGoogle Scholar
  2. 2.
    Lowe, D.: Distinctive image features from scale-invariant keypoints. Int. Journal on Computer Vision 60, 91–110 (2004)CrossRefGoogle Scholar
  3. 3.
    Baya, H., Essa, A., Tuytelaarsb, T., Gool, L.V.: Speed up robust features (surf). Computer Vision and Image Understanding 110(3), 346–359 (2008)CrossRefGoogle Scholar
  4. 4.
    Torr, P., Zisserman, A.: Mlesac: a new robust estimator with application to estimating image geometry. Computer Vision and Image Understanding 78(1), 138–156 (2000)CrossRefGoogle Scholar
  5. 5.
    Fischler, M.A., Bolles, R.: Random sample consensus: a paradigm for model fitting with applications to image analysis and automated cartography. Comm. ACM 24(6), 381–395 (1981)MathSciNetCrossRefGoogle Scholar
  6. 6.
    Fleck, D., Duric, Z.: Affine invariant-based classification of inliers and outliers for image matching. In: Kamel, M., Campilho, A. (eds.) ICIAR 2009. LNCS, vol. 5627, pp. 268–277. Springer, Heidelberg (2009)CrossRefGoogle Scholar
  7. 7.
    Pajares, G., Cruz, J.: Local stereovision matching through the adaline neural network. Pattern Recognition Letters 22, 1457–1473 (2001)CrossRefMATHGoogle Scholar
  8. 8.
    Pajares, G., Cruz, J.: Stereovision matching through support vector machines. Pattern Recognition Letters 24, 2575–2583 (2003)CrossRefGoogle Scholar
  9. 9.
    Foresti, G., Pieroni, G.: Exploiting neural trees in range image understanding. Pattern Recognit. Lett. 19 (9), 869–878 (1996)CrossRefGoogle Scholar
  10. 10.
    Mikolajczyk, K., Schmid, C.: Indexing based on scale invariant interest points. In: Int. Conf. on Computer Vision, vol. 1, pp. 525–531 (2001)Google Scholar
  11. 11.
    Utgoff, P.E.: Perceptron tree: A case study in hybrid concept representation. Connection Science 1(4), 377–391 (1989)CrossRefGoogle Scholar
  12. 12.
    Sankar, A., Mammone, R.: Neural Tree Networks. In: Neural Network: Theory and Application, pp. 281–302. Academic Press Professional, Inc., San Diego (1992)Google Scholar
  13. 13.
    Hall, M., Frank, E., Holmes, G., Pfahringer, B., Reutemann, P., Witten, I.: The weka data mining software: An update. SIGKDD Explorations 11(1) (2009)Google Scholar
  14. 14.
    Freund, Y., Schpire, R.: Experiments with a new boosting algorithm. In: Int. Conf. on Machine Learning, pp. 148–156. Morgan Kaufmann Pub. Inc., San Francisco (1996)Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2011

Authors and Affiliations

  • Sanjeev Kumar
    • 1
  • Asha Rani
    • 2
  • Christian Micheloni
    • 2
  • Gian Luca Foresti
    • 2
  1. 1.Department of MathematicsIIT RoorkeeRoorkeeIndia
  2. 2.Department of Mathematics and Computer ScienceUniversity of UdineUdineItaly

Personalised recommendations