Advertisement

Matching object models to segments from an optical flow field

  • Henner Kollnig
  • Hans-Hellmut Nagel
Conference paper
Part of the Lecture Notes in Computer Science book series (LNCS, volume 1065)

Abstract

The temporal changes of gray value structures recorded in an image sequence contain significantly more information about the recorded scene than the gray value structures of a single image. By incorporating optical flow estimates into the measurement function, our 3D pose estimation process exploits interframe information from an image sequence in addition to intraframe aspects used in previously investigated approaches. This increases the robustness of our vehicle tracking system and facilitates the correct tracking of vehicles even if their images are located in low contrast image areas. Moreover, partially occluded vehicles can be tracked without modeling the occlusion explicitly. The influence of interframe and intraframe image sequence data on pose estimation and vehicle tracking is discussed systematically based on various experiments with real outdoor scenes.

Keywords

Image Sequence Optical Flow Image Gradient Pixel Location World Coordinate System 
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. [Bar-Shalom & Fortmann 88]
    Y. Bar-Shalom, T. E. Fortmann, Tracking and Data Association, Academic Press, Inc., Boston/MA, Orlando/FL, and others, 1988Google Scholar
  2. [Barron et al. 94]
    J.L. Barron, D.J. Fleet, and S.S. Beauchemin, Performance of Optical Flow Techniques, International Journal of Computer Vision 12 (1994) 43–77Google Scholar
  3. [Bouthemy & FranÇois 93]
    P. Bouthemy, E. FranÇois, Motion Segmentation and Qualitative Dynamic Scene Analysis from an Image Sequence, International Journal of Computer Vision 10 (1993) 157–182.Google Scholar
  4. [Cédras & Shah 95]
    C. Cédras, M. Shah, Motion-Based Recognition: A Survey, Image and Vision Computing 13:2 (1995) 129–155.Google Scholar
  5. [Gelb 74]
    A. Gelb (ed.), Applied Optimal Estimation, MIT Press, Cambridge/MA, 1974.Google Scholar
  6. [Gong & Buxton 93]
    S. Gong, H. Buxton, From Contextual Knowledge to Computational Constraints, in Proc. British Machine Vision Conference, Guildford/UK, Sept. 21–23, 1993, pp. 229–238.Google Scholar
  7. [Koller et al. 93]
    D. Koller, K. Daniilidis, H.-H. Nagel, Model-Based Object Tracking in Monocular Image Sequences of Road Traffic Scenes, International Journal of Computer Vision 10 (1993) 257–281.Google Scholar
  8. [Kollnig et al. 94]
    H. Kollnig, H.-H. Nagel, and M. Otte, Association of Motion Verbs with Vehicle Movements Extracted from Dense Optical Flow Fields, in J.-O. Eklundh (ed.), Proc. Third European Conference on Computer Vision ECCV '94, Vol. II, Stockholm, Sweden, May 2–6, 1994, Lecture Notes in Computer Science 801, Springer-Verlag, Berlin, Heidelberg, New York/NY, and others, 1994, pp. 338–347.Google Scholar
  9. [Kollnig & Nagel 95]
    H. Kollnig and H.-H. Nagel, 3D Pose Estimation by Fitting Image Gradients Directly to Polyhedral Models, Proc. Fifth International Conference on Computer Vision ICCV '95, Cambridge/MA, 20–23 June 1995, pp. 569–574Google Scholar
  10. [Lowe 87]
    D.G. Lowe, Three-Dimensional Object Recognition from Single Two-Dimensional Images, Artificial Intelligence 31 (1987) 355–395.Google Scholar
  11. [Otte & Nagel 94]
    M. Otte, H.-H. Nagel, Optical Flow Estimation: Advances and Comparisons, Proc. Third European Conference on Computer Vision ECCV '94, Vol. I, Stockholm / Sweden, 2–6 May 1994, J.-O. Eklundh (ed.), Lecture Notes in Computer Science 800, Springer-Verlag Berlin Heidelberg New York/NY 1994, pp. 51–60.Google Scholar
  12. [Otte & Nagel 95]
    M. Otte, H.-H. Nagel, Estimation of Optical Flow Based on Higher-Order Spatiotemporal Derivatives in Interlaced and Non-Interlaced Image Sequences, Artificial Intelligence 78 (1995) 5–43Google Scholar
  13. [Proesmans et al. 94]
    M. Proesmans, L. Van Gool, E. Pauwels, A. Oosterlinck, Determination of Optical Flow and Its Discontinuities Using Non-Linear Diffusion, in J.-O. Eklundh (Ed.), Proc. Third European Conference on Computer Vision ECCV '94, Vol. II, Stockholm, Sweden, May 2–6, 1994, Lecture Notes in Computer Science 801, Springer-Verlag, Berlin, Heidelberg, New York/NY and others, 1994, pp. 295–304.Google Scholar
  14. [Schirra et al. 87]
    J.R.J. Schirra, G. Bosch, C.K. Sung, G. Zimmermann, From Image Sequences to Natural Language: A First Step towards Automatic Perception and Description of Motion, Applied Artificial Intelligence 1 (1987) 287–307.Google Scholar
  15. [Sullivan 92]
    G.D. Sullivan, Visual Interpretation of Known Objects in Constrained Scenes, Philosophical Transactions Royal Society London (B) 337 (1992) 361–370.Google Scholar
  16. [Sullivan et al. 95]
    G.D. Sullivan, A.D. Worrall, and J.M. Ferryman, Visual Object Recognition Using Deformable Models of Vehicles, in Proc. Workshop on Context-Based Vision, 19 June 1995, Cambridge/MA, pp. 75–86.Google Scholar
  17. [Tan et al. 94]
    T.N. Tan, G.D. Sullivan, K.D. Baker, Fast Vehicle Localization and Recognition without Line Extraction and Matching, in Proc. British Machine Vision Conference, York/UK, Sept. 13–16, 1994, pp. 85–94.Google Scholar
  18. [Worrall et al. 94]
    A.D. Worrall, G.D. Sullivan, and K.D. Baker, Pose Refinement of Active Models Using Forces in 3D, in J.-O. Eklundh (ed.), Proc. Third European Conference on Computer Vision (ECCV '94), Vol. I, Stockholm, Sweden, May 2–6, 1994, Lecture Notes in Computer Science 800, Springer-Verlag, Berlin, Heidelberg, New York/NY, and others, 1994, pp. 341–350.Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 1996

Authors and Affiliations

  • Henner Kollnig
    • 1
  • Hans-Hellmut Nagel
    • 1
    • 2
  1. 1.Institut für Algorithmen und Kognitive SystemeFakultät für Informatik der Universität Karlsruhe (TH)KarlsruheGermany
  2. 2.Fraunhofer-Institut für Informations- und Datenverarbeitung (IITB)KarlsruheGermany

Personalised recommendations