Automatic Pose Estimation Using Contour Information from X-Ray Images

  • Erik Soltow
  • Bodo Rosenhahn
Conference paper
Part of the Lecture Notes in Computer Science book series (LNCS, volume 9555)


An automatic approach to model-based pose estimation is presented that only uses given, segmented contours in X-ray images. The pose estimation based on Fourier Descriptors is extended to an arbitrary, calibrated stereo setup and eliminates the need for a manually given initialization. To further refine the pose, local and global optimization schemes that minimize the distance between the segmentation and the projected object over the six pose parameters are compared. Experiments show that a sampling-based optimization outperforms gradient-based methods and the sampling can be further improved to fit the given imaging setup and the object of interest. Simulated data shows that the pose can be estimated nearly perfect in stereo setups and yields highly accurate results on single-view setups. Clinical data supports these findings.


Pose estimation Fourier descriptors Simulated Annealing 



We would like to thank the Laboratory for Biomechanics and Biomaterials, Dept. of Orthopaedics, Hannover Medical School for providing the models and the clinical data used in the evaluation.


  1. 1.
    Banks, S.A., Hodge, W.A.: Accurate measurement of three-dimensional knee replacement kinematics using single-plane fluoroscopy. IEEE Trans. Biomed. Eng. 43, 638–649 (1996)CrossRefGoogle Scholar
  2. 2.
    Davis, L.: Genetic Algorithms and Simulated Annealing. Morgan Kaufman Publishers Inc., Los Altos (1987)zbMATHGoogle Scholar
  3. 3.
    Gamage, P., Xie, S.Q., Delmas, P.: Pose estimation of femur fracture segments for image guided orthopedic surgery. In: 24th International Conference Image and Vision Computing New Zealand, pp. 288–292 (2009)Google Scholar
  4. 4.
    Kaptein, B., Valstar, E., Stoel, B., Rozing, P., Reiber, J.: A new model-based RSA method validated using CAD models and models from reversed engineering. J. Biomech. 36, 873–882 (2003)CrossRefGoogle Scholar
  5. 5.
    Li, G., de Velde, S.K.V., Bingham, J.T.: Validation of a non-invasive fluoroscopic imaging technique for the measurement of dynamic knee joint motion. J. Biomech. 41, 1616–1622 (2008)CrossRefGoogle Scholar
  6. 6.
    Markelj, P., Tomazevic, D., Likar, B., Pernus, F.: A review of 3D/2D registration methods for image guided interventions. Med. Image Anal. 16, 642–661 (2012)CrossRefGoogle Scholar
  7. 7.
    Zhang, B., Sun, J.W., Chi, Z.Y., Sun, S.B., Meng, S., Zhang, Y., Rolfe, P.: Estimation of 3D pose of femurs in bi-planar digital radiographs. In: Long, M. (ed.) World Congress on Medical Physics and Biomedical Engineering May 26-31, 2012 Beijing, IFMBE Proceedings, vol. 39, pp. 1313–1316. Springer, Heidelberg (2013)Google Scholar
  8. 8.
    Yamazaki, T., Kamei, R., Tomita, T., Sato, Y., Yoshikawa, H., Sugamoto, K.: Improved semi-automated 3D kinematic measurement of total knee arthroplasty using x-ray fluoroscopic images. In: Jaffray, D.A. (ed.) World Congress on Medical Physics and Biomedical Engineering, June 7-12, 2015, Toronto, Canada, IFMBE Proceedings, vol. 51, pp. 322–325. Springer, Heidelberg (2016)Google Scholar
  9. 9.
    Zhang, D., Lu, G.: A comparative study of fourier descriptors for shape representation and retrieval. In: Proceedings of 5th Asian Conference on Computer Vision (2002)Google Scholar
  10. 10.
    Nocedal, J., Wright, S.J.: Numerical Optimization. Springer, New York (2006)zbMATHGoogle Scholar
  11. 11.
    Wunsch, P., Hirzinger, G.: Registration of CAD-models to images by iterative inverse perspective matching. In: Proceedings of the 1996 International Conference on Pattern Recognition (1996)Google Scholar

Copyright information

© Springer International Publishing Switzerland 2016

Authors and Affiliations

  1. 1.Institut Für InformationsverarbeitungLeibniz Universität HannoverHannoverGermany

Personalised recommendations