Motion and structure factorization and segmentation of long multiple motion image sequences

  • Chris Debrunner
  • Narendra Ahuja
Conference paper
Part of the Lecture Notes in Computer Science book series (LNCS, volume 588)


This paper presents a computer algorithm which, given a dense temporal sequence of intensity images of multiple moving objects, will separate the images into regions showing distinct objects, and for those objects which are rotating, will calculate the three-dimensional structure and motion. The method integrates the segmentation of trajectories into subsets corresponding to different objects with the determination of the motion and structure of the objects. Trajectories are partitioned into groups corresponding to the different objects by fitting the trajectories from each group to a hierarchy of increasingly complex motion models. This grouping algorithm uses an efficient motion estimation algorithm based on the factorization of a measurement matrix into motion and structure components. Experiments are reported using two real image sequences of 50 frames each to test the algorithm.


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.


  1. [Ad1]
    Adiv, G.: Determining Three-Dimensional Motion and Structure from Optical Flow Generated by Several Moving Objects. IEEE Transactions on PAMI 7 (1985) 384–401Google Scholar
  2. [BB1]
    Bergen, J., Burt, P., Hingorani, R., Peleg, S.: Multiple Component Image Motion: Motion Estimation. Proc. of the 3d ICCV, Osaka, Japan (December 1990) 27–32Google Scholar
  3. [BB2]
    Boult, T., Brown, L.: Factorization-based Segmentation of Motions. Proc. of the IEEE Motion Workshop, Princeton NJ (October 1991) 21–28Google Scholar
  4. [BH1]
    Blostein, S., Huang, T.: Detecting Small, Moving Objects in Image Sequences using Sequential Hypothesis Testing. IEEE Trans. on Signal Proc. 39 (July 1991) 1611–1629Google Scholar
  5. [DA1]
    Debrunner, C., Ahuja, N.: A Direct Data Approximation Based Motion Estimation Algorithm. Proc. of the 10th ICPR, Atlantic City, NJ (June 1990) 384–389Google Scholar
  6. [DA2]
    Debrunner, C., Ahuja, N.: Estimation of Structure and Motion from Extended Point Trajectories. (submitted)Google Scholar
  7. [DA3]
    Debrunner, C., Ahuja, N.: Motion and Structure Factorization and Segmentation of Long Multiple Motion Image Sequences. (submitted)Google Scholar
  8. [De1]
    Debrunner, C.: Structure and Motion from Long Image Sequences. Ph.D. dissertation, University of Illinois at Urbana-Champaign, Urbana, IL (August 1990)Google Scholar
  9. [Ka1]
    Kanade, T.: personal communication (October 1991)Google Scholar
  10. [KA1]
    Kung, S., Arun, K., Rao, D.: State-Space and Singular-Value Decomposition-Based Approximation Methods for the Harmonic Retrieval Problem. J.of the Optical Society of America (December 1983) 1799–1811Google Scholar
  11. [TK1]
    Tomasi, C., Kanade, T.: Factoring Image Sequences into Shape and Motion. Proc. of the IEEE Motion Workshop, Princeton NJ (October 1991) 21–28Google Scholar
  12. [WH1]
    Weng, J., Huang, T., Ahuja, N.: Motion and Structure from Two Perspective Views: Algorithms, Error Analysis, and Error Estimation. IEEE Transactions on PAMI 11 (1989) 451–476Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 1992

Authors and Affiliations

  • Chris Debrunner
    • 1
  • Narendra Ahuja
    • 1
  1. 1.Coordinated Science LaboratoryUniversity of Illinois at Urbana-ChampaignUrbana

Personalised recommendations