Advertisement

Tracking with a Novel Pose Estimation Algorithm

  • Bodo Rosenhahn
  • Norbert Krüger
  • Torge Rabsch
  • Gerald Sommer
Conference paper
Part of the Lecture Notes in Computer Science book series (LNCS, volume 1998)

Abstract

In this paper we apply a novel pose estimation algorithm to the tracking problem. We make use of error measures of the algorithm which enable us to characterize the quality of an estimated pose. The key idea of the tracking algorithm is random start local search. The principle of the heuristic relies upon a combination of iterative improvement and random sampling. While in many approaches a manually designed object representation is assumed, we overcome this condition by using accumulated object representations and combine these successfully with the tracking algorithm.

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

References

  1. 1.
    M. Ackermann: Akkumulieren von Objektrepräsentationen im Wahrnehmungs-Handlungs Zyklus. Diplomarbeit,Le hrstuhl für Kognitive Systeme der Christian-Albrechts-Universität zu Kiel,( 2000). 11Google Scholar
  2. 2.
    E. Bayro-Corrochano: The geometry and algebra of kinematics. In: Geometric Computing with Clifford Algebra (Sommer G.,ed.), Springer, Berlin (2000). 12Google Scholar
  3. 3.
    P. V. C. Hough: Methods and means for recognizing complex patterns. U. S. Patent 3,069,654, (Dec. 18, 1962). 10Google Scholar
  4. 4.
    Norbert Krüger, Marcus Ackermann, Gerald Sommer: Accumulation of object representations utilizing interaction of robot action and perception. DAGM 2000 (2000) 365–372. 10, 15Google Scholar
  5. 5.
    Norbert Krüger, Daniel Wendor., Gerald Sommer: Two models of a vision-based robotic system: visual haptic attention and accumulation of object representations. In:Proceed. Robot Vision 2001,Auckland (2001). 15, 17Google Scholar
  6. 6.
    S. Kunze: Ein Hand-Auge-System zur visuell basierten Lokalisierung und Identifikation von Objekten. Diplomarbeit,Le hrstuhl für Kognitive Systeme der Christian-Albrechts-Universität zu Kiel, (2000). 11Google Scholar
  7. 8.
    J. Pauli: Geometric/photometric consensus and regular shape quasi-invariants for object localization and boundary extraction. Technical Report 9805, Christian-Albrechts-Universität zu Kiel,Ins titut für Informatik und Praktische Mathematik, (1998). 11Google Scholar
  8. 9.
    J. Princen, J. Illingworth, J. Kittler: An optimizing line finder using a Hough transform algorithm. Computer Vision, Graphics, and Image Processing 52 (1990) 57–77. 10CrossRefGoogle Scholar
  9. 10.
    B. Rosenhahn, Y. Zhang, G. Sommer: Pose estimation in the language of kinematics. 2nd Internat. Workshop Algebraic Frames for the Perception-Action Cycle, AFPAC 2000, LNCS 1888 (2000) 284–293. 11Google Scholar
  10. 11.
    J. R. Beveridge: Local search algorithms for geometric object recognition: Optimal correspondence and pose. Technical Report CS 93-5, University of Massachusetts, (1993). 10, 13Google Scholar
  11. 12.
    G. Sommer, B. Rosenhahn, Y. Zang: Pose estimation using geometric constraints. Technical Report 2003, Christian-Albrechts-Universität zu Kiel,Ins titut für Informatik und Praktische Mathematik,( 2000). 9, 10, 11, 13, 14, 16, 17Google Scholar
  12. 13.
    Y. Zhang, B. Rosenhahn, G. Sommer: Extended Kalman filter design for motion estimation by point and line observations. 2nd Internat.Workshop Algebraic Frames for the Perception-Action Cycle,A FPAC 2000, LNCS 1888 (2000) 339–348. 13Google Scholar
  13. 14.
    Y. Zhang, G. Sommer: 3D rigid object tracking in computer vision. Seminar presentation, Oberseminar, Institut für Informatik und Praktische Mathematik, Christian-Albrechts-Universität zu Kiel (July 2000). 14Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2001

Authors and Affiliations

  • Bodo Rosenhahn
    • 1
  • Norbert Krüger
    • 1
  • Torge Rabsch
    • 1
  • Gerald Sommer
    • 1
  1. 1.Institut für Informatik und Praktische MathematikChristian-Albrechts-Universität zu KielKielGermany

Personalised recommendations