New techniques for 3D modeling ... ... and for doing without

Chapter 3 Vision
Part of the Lecture Notes in Control and Information Sciences book series (LNCIS, volume 250)

Abstract

Object recognition, visual robot guidance, and several other vision applications require models of objects or scenes. Computer vision has a tradition of building these models from inherent object characteristics. The problem is that such characteristics are difficult to extract. Recently, a pure view-based object recognition approach was proposed, that is surprisingly performant. It is based on a model that is extracted directly from raw image data. Limitations of both strands raise the question whether there is room for middle ground solutions, that combine the strengths but avoid the weaknesses. Two examples are discussed, where in each case the only input required are images, but where nevertheless substantial feature extraction and analysis are involved. These are non-Euclidean 3D reconstruction from multiple, uncalibrated views and scene description based on local, affinely invariant surface patches that can be extracted from single views. Both models are useful for robot vision tasks such as visual navigation.

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

References

  1. [1]
    Murase H, Nayar S 1995 Visual learning and recognition of 3-D objects from appearance. International Journal of Computer Vision 14:5–24CrossRefGoogle Scholar
  2. [2]
    Moons T 1998 A guided tour through multiview relations. In: Koch R, Van Gool L (eds) 1998 Proceedings of the SMILE Workshop — 3D Structure from Multiple Images of Large-scale Environments. Springer-Verlag, Berlin (Lecture Notes in Computer Science No. 1506), pp. 304–346CrossRefGoogle Scholar
  3. [3]
    Armstrong M, Zisserman A, Beardsley P 1994 Euclidean structure from uncalibrated images. Proc. British Machine Vision Conference (BMVC’ 94) pp.Google Scholar
  4. [4]
    Pollefeys M, Koch R, Van Gool L 1998 Self-calibration and metric reconstruction in spite of varying and unknown internal camera parameters. Proc. International Conference on Computer Vision (ICCV’ 98), Bombay, India, pp. 90–95Google Scholar
  5. [5]
    Triggs B 1997 The absolute quadric. Proc. International Conference on Computer Vision and Pattern Recognition (ICCV’ 97), pp. 609–614Google Scholar
  6. [6]
    Heyden A and Åström K 1997 Euclidean reconstruction from image sequences with varying and unknown focal length and principal point. Proc. IEEE Conference on Computer Vision and Pattern Recognition (CVPR’ 97) Google Scholar
  7. [7]
    Pollefeys M, Van Gool L, Proesmans M 1996 Euclidean 3D reconstruction from image sequences with variable focal lengths. Proc. European Conference on Computer Vision (ECCV’ 96) pp. 31–42Google Scholar
  8. [8]
    Moons T, Van Gool L, Proesmans M, Pauwels E 1996 Affine reconstruction from perspective image pairs with a relative object-camera translation in between. IEEE Trans. Pattern Analysis and Machine Intelligence (T-PAMI) 18:77–83CrossRefGoogle Scholar
  9. [9]
    Mundy J, Zisserman A (eds) 1992 Applications of invariance in vision. MIT Press, BostonGoogle Scholar
  10. [10]
    Schmid C, Mohr R 1997 Local greyvalue invariants for image retrieval. IEEE Trans. Pattern Analalysis and Machine Intelligence (T-PAMI) 19:872–877Google Scholar
  11. [11]
    Pritchett P, Zisserman A 1998 Wide baseline stereo matching. Proc. International Conference on Computer Vision (ICCV’ 98), pp. 754–759Google Scholar
  12. [12]
    Mindru F, Moons T, Van Gool L 1998 Color-based moment invariants for the viewpoint and illumination independent recognition of planar color patterns, Proc. International Conference on Application of Pattern Recognition (ICAPR’ 98), pp. 113–122Google Scholar
  13. [13]
    Tuytelaars T, Van Gool L, D’Haene L, Koch R 1999 Matching affinely invariant regions for visual servoing, accepted for oral presentation at the International Conference on Robotics and Automation, DetroitGoogle Scholar
  14. [14]
    Shade J, Gortler S, He L-W, Szeliski R 1998 Layered Depth Images. Computer Graphics (SIGGRAPH’ 98) Google Scholar
  15. [15]
    Debevec P E, Taylor C J, Malik J 1996 Modeling and rendering architecture from photographs: A hybrid geometry-and image-based approach. Computer Graphics (SIGGRAPH’ 96), pp. 11–20Google Scholar

Copyright information

© Springer-Verlag London Limited 2000

Authors and Affiliations

  1. 1.ESAT, Univ. of LeuvenLeuvenBelgium

Personalised recommendations