Advertisement

Analysis of scenes containing multiple non-polyhedral 3D objects

  • Mauro S. Costa
  • Linda G. Shapiro
Scene Understanding
Part of the Lecture Notes in Computer Science book series (LNCS, volume 974)

Abstract

Recognition of generic three-dimensional objects remains an unsolved problem. Scenes containing multiple nonpolyhedral 3D objects are particularly challenging. Conventional object models based on straight line segments and junctions are not suitable for this task. We have developed an appearanced-based 3D object model in which an object is represented by the features that can be most reliably detected in a training set of real images. For industrial objects with both flat and curved surfaces, holes, and threads, a set of useful features has been derived; and a recognition system utilizing these features and their interrelationships is being developed. The recognition system uses small relational subgraphs of features to index the database of models and to retrieve the appropriate 3D models in a hypothesize- and-test matching algorithm. This paper describes the new models, the matching algorithm, and our preliminary results.

References

  1. 1.
    R. Bergevin and M. D. Levine. Generic Object Recognition: Building and Matching Coarse Descriptions from Line Drawings. IEEE Transactions on Pattern Analysis and Machine Intelligence, 15(1):19–36, 1993.Google Scholar
  2. 2.
    R. C. Bolles and R. A. Cain. Recognizing and Locating Partially Visible Objects: The Local-Feature-Focus Method. The International Journal of Robotics Research, 1(3):57–82,1982.Google Scholar
  3. 3.
    O. I. Camps, L. G. Shapiro, and R. M. Haralick. Image Prediction for Computer Vision. In Three-dimensional Object Recognition Systems, A. Jain and P. Flynn (eds). Elsevier Science Publishers BV, 1993.Google Scholar
  4. 4.
    M. S. Costa, R. M. Haralick and L. G. Shapiro. Optimal Affine Invariant Point Matching. In Proceedings of 10th ICPR, volume 1, pp. 233–236, 1990.Google Scholar
  5. 5.
    A. Etemadi. Robust segmentation of edge data. In Proceedings of the IEE Image Processing Conference, 1992.Google Scholar
  6. 6.
    K. D. Gremban and K. Ikeuchi. Appearance-Based Vision and the Automatic Generation of Object Recognition Programs. In Three-dimensional Object Recognition Systems, A. Jain and P. Flynn (eds). Elsevier Science Publishers BV, 1993.Google Scholar
  7. 7.
    R. Hummel and H. Wolfson. Affine Invariant Matching. DARPA Image Understanding Workshop, April, 1988.Google Scholar
  8. 8.
    A. K. Jain and R. Hoffman. Evidence-Based Recognition of 3-D Objects. IEEE Transactions on Pattern Analysis and Machine Intelligence, 10(6):783–802, 1988.Google Scholar
  9. 9.
    H. Murase and S. K. Nayar. Visual Learning of Object Models from Appearance. International Journal of Computer Vision, in press. Also Tech. Rep. CUCS-054-92.Google Scholar
  10. 10.
    K. Pulli. TRIBORS: A Triplet-Based Object Recognition System. Technical Report 95-01-01, Department of Computer Science and Engineering, University of Washington, January 1995.Google Scholar
  11. 11.
    L. G. Shapiro and M. S. Costa. Appearance-Based 3D Object Recognition. In Proc. of the NSF/DARPA Workshop on 3D Object Representation for Computer Vision, New York, NY, December 1994.Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 1995

Authors and Affiliations

  • Mauro S. Costa
    • 1
  • Linda G. Shapiro
    • 1
    • 2
  1. 1.Department of Electrical EngineeringUniversity of WashingtonSeattleUSA
  2. 2.Department of Computer Science & EngineeringUniversity of WashingtonSeattleUSA

Personalised recommendations