Multiple Hypothesis Target Tracking Using Merge and Split of Graph’s Nodes

  • Yunqian Ma
  • Qian Yu
  • Isaac Cohen
Part of the Lecture Notes in Computer Science book series (LNCS, volume 4291)


In this paper, we propose a maximum a posteriori formulation to the multiple target tracking problem. We adopt a graph representation for storing the detected regions as well as their association over time. The multiple target tracking problem is formulated as a multiple paths search in the graph. Due to the noisy foreground segmentation, an object may be represented by several foreground regions and one foreground region may corresponds to multiple objects. We introduce merge, split and mean shift operations that add new hypothesis to the measurement graph in order to be able to aggregate, split detected blobs or re-acquire objects that have not been detected during stop-and-go-motion. To make full use of the visual observations, we consider both motion and appearance likelihood. Experiments have been conducted on both indoor and outdoor data sets, and a comparison has been carried to assess the contribution of the new tracker.


Tracking Algorithm Appearance Model Shift Operation Foreground Region Probabilistic Data Association 
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.


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.


  1. 1.
    LaScala, B., Pulford, G.W.: Viterbi data association tracking for over-the-horizon radar. In: Proc. Int. Radar Sym., September 1998, vol. 3, pp. 155–164 (1998)Google Scholar
  2. 2.
    Buckley, K., Vaddiraju, A., Perry, R.: A new pruning/merging algorithm for MHT multitarget tracking, Radar–2000 (May 2000)Google Scholar
  3. 3.
    Quach, T., Farooq, M.: Maximum likelihood track formation with the Viterbi Algorithm. In: Proc. 33-rd IEEE Conf. on Decision and Control, pp. 271–276 (December 1994)Google Scholar
  4. 4.
    Castanon, D.: Efficient algorithms for finding the K best paths through a trellis. IEEE Trans. on Aerospace & Elect. Sys. 26(2), 405–410 (1990)CrossRefGoogle Scholar
  5. 5.
    Fortman, T.E., Bar-Shalom, Y., Scheffe, M.: Sonar tracking of Multiple Targets Using Joint Probabilistic Data Association. IEEE Journal of Oceanic Engineering OE-8(3), 173–184 (1983)CrossRefGoogle Scholar
  6. 6.
    Bar-Shalom, Y., Fortmann, T.E.: Tracking and Data Association. Mathematics in Science and Engineering Series, vol. 179. Academic Press, San Diego (1988)MATHGoogle Scholar
  7. 7.
    Reid, D.B.: An algorithm for tracking multiple targets. IEEE Transaction on Automatic Control 24(6), 843–854 (1979)CrossRefGoogle Scholar
  8. 8.
    Cox, I.J., Hingorani, S.L.: An Efficient Implementation of Reid’s Multiple Hypothesis Tracking Algorithm and Its Evaluation for the Purpose of Visual Tracking. PAMIGoogle Scholar
  9. 9.
    Kang, J., Cohen, I., Medioni, G.: Object reacquisition using invariant appearance model. In: ICPR, pp. 759–762, 18–2,138–150 (1996) (2004)Google Scholar
  10. 10.
    Kang, J., Cohen, I., Medioni, G.: Continuous Tracking Within and across Camera Streams. In: IEEE, Conference on CVPR 2003, Madison, Wisconsin (2003)Google Scholar
  11. 11.
    Kang, J., Cohen, I., Medioni, G.: Persistent Objects Tracking Across Multiple Non Overlapping Cameras. In: IEEE Workshop on Motion and Video Computing, MOTION 2005., Breckenridge, Colorado (January 4-5, 2005)Google Scholar
  12. 12.
    Yizong, C.: Mean Shift, Mode Seeking, and Clustering. PAMI 17(8), 790–799 (1995)Google Scholar
  13. 13.
    Blackman, S.: Multiple Target Tracking with Radar Applications. Artech House (1986)Google Scholar
  14. 14.
    Bar-Shalom, Y., Tse, E.: Tracking in a Cluttered Environment with Probabilistic Data Association. Automatica, 451–460 (1975)Google Scholar
  15. 15.
    Genovesio, A., Olivo-Marin, J.: Split and merge data association filter for dense multi-target tracking. In: ICPR, vol. IV, pp. 677–680 (2004)Google Scholar
  16. 16.
    Senior, A.: Tracking People with Probabilistic Appearance Models. In: PETS 2002, pp. 48–55 (2002)Google Scholar
  17. 17.
    Ying, W., Ting, Y., Gang, H.: Tracking Appearances with Occlusions. In: Proc. IEEE Conf. on CVPR 2003, Madison, vol. I, pp. 789–795 (June 2003)Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2006

Authors and Affiliations

  • Yunqian Ma
    • 1
  • Qian Yu
    • 2
  • Isaac Cohen
    • 1
  1. 1.Honeywell LabsMinneapolis
  2. 2.Institue for Robotics and Intelligent SystemsUniversity of Southern California 

Personalised recommendations