Advertisement

Recognition by Probabilistic Hypothesis Construction

Conference paper
Part of the Lecture Notes in Computer Science book series (LNCS, volume 3021)

Abstract

We present a probabilistic framework for recognizing objects in images of cluttered scenes. Hundreds of objects may be considered and searched in parallel. Each object is learned from a single training image and modeled by the visual appearance of a set of features, and their position with respect to a common reference frame. The recognition process computes identity and position of objects in the scene by finding the best interpretation of the scene in terms of learned objects. Features detected in an input image are either paired with database features, or marked as clutters. Each hypothesis is scored using a generative model of the image which is defined using the learned objects and a model for clutter. While the space of possible hypotheses is enormously large, one may find the best hypothesis efficiently – we explore some heuristics to do so. Our algorithm compares favorably with state-of-the-art recognition systems.

Keywords

Model Feature Training Image Constellation Model Probabilistic Framework Scene Image 
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.
    Clarke, F., Ekeland, I.: Nonlinear oscillations and boundary-value problems for Hamiltonian systems. Arch. Rat. Mech. Anal. 78, 315–333 (1982)zbMATHCrossRefMathSciNetGoogle Scholar
  2. 2.
    Schiele, B.: Object Recognition Using Multidimensional Receptive Field Histograms PhD thesis, I.N.P. de Grenoble (1997)Google Scholar
  3. 3.
    Schmid, C., Mohr, R.: Local greyvalue invariants for image retrieval. IEEE Trans. on Patt. Anal. Mach. Int. 19(5), 530–535 (1997)CrossRefGoogle Scholar
  4. 4.
    Burl, M.C., Perona, P.: Recognition of planar object classes. In: IEEE Comp. on Comp. Vision and Patt. Recog., CVPR 1996, San Francisco, CA (June 1996)Google Scholar
  5. 5.
    Fergus, R., Perona, P., Zisserman, A.: Object class recognition by unsupervised scale-invariant learning. In: IEEE Conf. on Comp. Vision and Patt. Recog. (2003)Google Scholar
  6. 6.
    Weber, M., Welling, M., Perona, P.: Unsupervised learning of models for recognition. In: Vernon, D. (ed.) ECCV 2000. LNCS, vol. 1842, pp. 18–32. Springer, Heidelberg (2000)CrossRefGoogle Scholar
  7. 7.
    Lowe, D.G.: Object recognition from local scale-invariant features. In: Proc. Int. Conf. Comp. Vision, Corfu, Greece, pp. 1150–1157 (1999)Google Scholar
  8. 8.
    Lowe, D.G.: Distinctive image features from scale-invariant keypoints accepted paper. Int. J. of Comp. Vision (2004)Google Scholar
  9. 9.
    Weber, M.: Unsupervised Learning of Models for Object Recognition, Ph.D thesis, Department of Computation and Neural Systems, California Institute of Technology, Pasadena, CA (2000)Google Scholar
  10. 10.
    Coughlan, J.M., Yuille, A.L.: Bayesian A* tree search with expected O(N) node expansions: applications to road tracking. Draft submitted to Neural Computation (December 2002)Google Scholar
  11. 11.
    Pearl, J.: Heuristics. Addison-Wesley, Reading (1984)Google Scholar
  12. 12.
    Beis, J.S., Lowe, D.G.: Shape indexing using approximate nearest-neighbour search in high-dimensional spaces. In: Proc. IEEE Conf. on Comp. Vision and Patt. Recog., Puerto Rico, pp. 1000–1006 (1997)Google Scholar
  13. 13.
    Fei-Fei, L., Fergus, R., Perona, P.: A bayesian approach to unsupervised one-shot learning of object categories. In: Proc. Int. Conf. on Comp. Vision, Nice, France (2003)Google Scholar
  14. 14.
    Grimson, W.E.L.: Model-based recognition and localization from sparse range or tactile data, AI Memo 738, Massachusetts Institute of Technology (August 1983)Google Scholar
  15. 15.
    Mikolajczyk, K., Schmid, C.: A performance evaluation of local descriptors. In: Proc. IEEE Conf. on Comp. Vision and Patt. Recog., Madison, Wisconsin, p. 257Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2004

Authors and Affiliations

  1. 1.California Institute of TechnologyPasadenaUSA

Personalised recommendations