Global Motion Model for Stereovision-Based Motion Analysis

  • Jia WangEmail author
  • Zhencheng Hu
  • Keiichi Uchimura
  • Hanqing Lu
Open Access
Research Article


An advantage of stereovision-based motion analysis is that the depth information is available, thus motion can be estimated more precisely in Open image in new window D stereo coordinate system (SCS) constructed by the depth and the image coordinates. In this paper, stereo global motion in SCS, which is induced by 3D camera motion in real-world coordinate system (WCS), is parameterized by a five-parameter global motion model (GMM). Based on such model, global motion can be estimated and identified directly in SCS without knowing the physical parameters about camera motion and camera setup in WCS. The reconstructed global motion field accords with the spatial structure of the scene much better. Experiments on both synthetic data and real-world images illustrate its promising performance.


Coordinate System Information Technology Physical Parameter Spatial Structure Quantum Information 


  1. 1.
    Huang TS, Tsai RY: Three-dimensional motion estimation from image-space shifts. Proceedings of the IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP '81), March–April 1981, Atlanta, Ga, USA 6: 1136–1139.CrossRefGoogle Scholar
  2. 2.
    Rath GB, Makur A: Iterative least squares and compression based estimations for a four-parameter linear global motion model and global motion compensation. IEEE Transactions on Circuits and Systems for Video Technology 1999, 9(7):1075–1099. 10.1109/76.795060CrossRefGoogle Scholar
  3. 3.
    Qian G, Chellappa R, Zheng Q: Robust Bayesian cameras motion estimation using random sampling. International Conference on Image Processing (ICIP '04), October 2004, Singapore, Republic of Singapore 2: 1361–1364.Google Scholar
  4. 4.
    Huang TS, Blostein SD: Robust algorithm for motion estimation based on two sequential stereo image pairs. Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition (CVPR '85), June 1985, San Francisco, Calif, USA 518–523.Google Scholar
  5. 5.
    Young G-SJ, Chellappa R: 3-D motion estimation using a sequence of noisy stereo images: models, estimation, and uniqueness results. IEEE Transactions on Pattern Analysis and Machine Intelligence 1990, 12(8):735–759. 10.1109/34.57666CrossRefGoogle Scholar
  6. 6.
    Shieh J, Zhuang H, Sudhakar R: Motion esimtation from a sequence of stereo images: a direct method. IEEE Transactions on Systems, Man, and Cybernetics 1994, 24(7):1044–1053. 10.1109/21.297792CrossRefGoogle Scholar
  7. 7.
    Hirschmuller H, Innocent PR, Garibaldi JM: Fast, unconstrained camera motion estimation from stereo without tracking and robust statistics. The 7th International Conference on Control, Automation, Robotics and Vision, December 2002, Singapore, Republic of Singapore 2: 1099–1104.Google Scholar
  8. 8.
    Xiang Z, Genc Y: Bootstrapped real-time ego motion estimation and scene modeling. The 5th International Conference on 3-D Digital Imaging and Modeling (3DIM '05), June 2005, Ottawa, Ontario, Canada 514–521.CrossRefGoogle Scholar
  9. 9.
    Hu Z, Uchimura K: U-V-disparity: an efficient algorithm for stereovision based scene analysis. IEEE Intelligent Vehicle Symposium (IV '05), June 2005, Las Vegas, Nev, USAGoogle Scholar
  10. 10.
    Tekalp AM: Digital Video Processing. Prentice Hall, Beijing, China; 1998.Google Scholar
  11. 11.
    Shi J, Tomasi C: Good features to track. Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition (CVPR '94), June 1994, Seattle, Wash, USA 593–600.Google Scholar
  12. 12.
    Kang SB, Szeliski R, Chai J: Handling occlusions in dense multi-view stereo. Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition (CVPR '01), December 2001, Kauai, Hawaii, USA 1: 103–110.Google Scholar

Copyright information

© Wang et al. 2006

Authors and Affiliations

  • Jia Wang
    • 1
    Email author
  • Zhencheng Hu
    • 2
  • Keiichi Uchimura
    • 2
  • Hanqing Lu
    • 1
  1. 1.National Laboratory of Pattern RecognitionInstitute of Automation, Chinese Academy of SciencesBeijingChina
  2. 2.Department of Computer Science, Faculty of EngineeringKumamoto UniversityKumamotoJapan

Personalised recommendations