On token-matching in real time motion analysis

  • Henrik I Christensen
  • Erik Granum
Motion And Depth Analysis
Part of the Lecture Notes in Computer Science book series (LNCS, volume 301)


A token matching scheme for high-level motion analysis has been developed. The goal is real-time usage for-a constrained scenario. The demand for real-time has forced an unsophisticated method, which simply propagates certainties through normalizations.

Matching has been performed between a measured and an estimate of the same frame. The experiments has shown that the matching method is capable of handling occlusions even in the case of rather poor estimates. The problems observed for very poor estimation, are clearly due to lack of sophistication in the estimators usage of motion information. In other words we have a number of problems to attend to and solve before the weaknesses of the matching scheme will appear.


Dynamic scene analysis robot vision image sequences probabilistic relaxation Support functions 


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.


  1. Christensen, H.I., 1987 Monitoring Moving Objects in Real Time, M.Sc Thesis, Institute of Electronic Systems, Aalborg University, Denmark.Google Scholar
  2. Goshtasby, A. & Page, C.V., 1984 Image matching by a probabislitic relaxation labeling process, Proceedings from Seventh International Conference on Pattern Recognition, pp. 307.Google Scholar
  3. Granum, E. & Munk, K. H., 1986 Monitoring moving objects in real time, Description of FTU-project, Institute of Electronic Systems, Alborg University, Denmark.Google Scholar
  4. Hummel, R.A. & Zucker, S.W., 1983 On the foundation of relaxation labeling processes, IEEE PAMI-5, No. 3, May 1983, pp. 267.Google Scholar
  5. Kittler, J. & Illingworth, J., 1985 Relaxation Labeling algorithms — a review, Image and Vision Computing, Vol 3. 1985, pp. 206CrossRefGoogle Scholar
  6. Munk, K.H., 1986 Bevaegelsesanalyse ved hojniveau billedrepraesentation, (Motion Analysis using High Level Image description), M.Sc Thesis (in Danish), Institute of Electronic Systems, Aalborg University, Denmark.Google Scholar
  7. Peleg, S., 1980 A new probabilistic relaxation scheme, IEEE PAMI-2, No. 4, July 1980, pp. 363.Google Scholar
  8. Price, K.E., 1985 Relaxation matching techniques — A Comparison, IEEE PAMI-7, No. 5, September 1985, pp. 617.Google Scholar
  9. Rosenfeld et. al, 1976 Scene labeling be relaxation operators, IEEE SMC-6, No. 6, June 1976, pp. 421.Google Scholar
  10. Sethi, I.K. & Jain, R., 1987 Finding Trajectories of Feature Points in A Monocular Image Sequence IEEE PAMI-9, No. 1, Januar 1987, pp. 56Google Scholar
  11. Ullman, S., 1979 Interpretation of visual motion, MIT Press, Massachusetts, USA.Google Scholar
  12. Wiklund, J. & Granlund, G.H., 1987 Image Sequence Analysis for Object Tracking, 5th Scandinavian Conference on Image Analysis, June 1987, pp. 641Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 1988

Authors and Affiliations

  • Henrik I Christensen
    • 1
  • Erik Granum
    • 1
  1. 1.Laboratory of Image Analysis, Institute of Electronic SystemsAalborg UniversityDenmark

Personalised recommendations