Shape-from-Silhouette with Two Mirrors and an Uncalibrated Camera

  • Keith Forbes
  • Fred Nicolls
  • Gerhard de Jager
  • Anthon Voigt
Part of the Lecture Notes in Computer Science book series (LNCS, volume 3952)


Two planar mirrors are positioned to show five views of an object, and snapshots are captured from different viewpoints. We present closed form solutions for calculating the focal length, principal point, mirror and camera poses directly from the silhouette outlines of the object and its reflections. In the noisy case, these equations are used to form initial parameter estimates that are refined using iterative minimisation. The self-calibration allows the visual cones from each silhouette to be specified in a common reference frame so that the visual hull can be constructed. The proposed setup provides a simple method for creating 3D multimedia content that does not rely on specialised equipment. Experimental results demonstrate the reconstruction of a toy horse and a locust from real images. Synthetic images are used to quantify the sensitivity of the self-calibration to quantisation noise. In terms of the silhouette calibration ratio, degradation in silhouette quality has a greater effect on silhouette set consistency than computed calibration parameters.


Focal Length Closed Form Solution Real Image Virtual Object Synthetic Image 
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.


  1. 1.
    Georgescu, B., Shimshoni, I., Meer, P.: Mean shift based clustering in high dimensions: A texture classification example. In: ICCV, pp. 456–463 (2003)Google Scholar
  2. 2.
    Boyer, E.: Camera calibration using silhouettes. Technical Report 5559, INRIA (2005)Google Scholar
  3. 3.
    Cipolla, R., Giblin, P.: Visual Motion of Curves and Surfaces. Cambridge University Press, Cambridge (2000)zbMATHGoogle Scholar
  4. 4.
    Forbes, K., Voigt, A., Bodika, N.: Visual hulls from single uncalibrated snapshots using two planar mirrors. In: Proc. 15th South African Workshop on Pattern Recognition (2004)Google Scholar
  5. 5.
    Gluckman, J.M., Nayar, S.K.: Planar catadioptric stereo: Geometry and calibration. In: Proc. IEEE Conference on Computer Vision and Pattern Recognition (1999)Google Scholar
  6. 6.
    Hu, B., Brown, C.M., Nelson, R.C.: Multiple-view 3-D reconstruction using a mirror. Technical Report TR863, Computer Science Dept., U. Rochester (May 2005)Google Scholar
  7. 7.
    Laurentini, A.: The visual hull concept for silhouette-based image understanding. IEEE Transactions on Pattern Analysis and Machine Intelligence 16(2), 150–162 (1994)CrossRefGoogle Scholar
  8. 8.
    Matusik, W., Buehler, C., McMillan, L.: Polyhedral visual hulls for real-time rendering. In: Proceedings of Twelfth Eurographics Workshop on Rendering (2001)Google Scholar
  9. 9.
    Moriya, T., Beniyama, F., Matsumoto, K., Minakawa, T., Utsugi, K., Takeda, H.: Properties of three shadows on a plane. In: Proceedings of the 12th International Conference in Central Europe on Computer Graphics, Visualization and Computer Vision (2004)Google Scholar
  10. 10.
    Mülayim, A., Yilmaz, U., Atalay, V.: Silhouette-based 3D model reconstruction from multiple images. IEEE Transactions on Systems, Man and Cybernetics 33(4) (2003)Google Scholar
  11. 11.
    Okatani, T., Deguchi, K.: Recovering camera motion from image sequence based on registration of silhouette cones. In: IAPR Workshop on Machine Vision Applications (2000)Google Scholar
  12. 12.
    Sinha, S., Pollefeys, M., McMillan, L.: Camera network calibration from dynamic silhouettes. In: Proc. IEEE Conference on Computer Vision and Pattern Recognition (2004)Google Scholar
  13. 13.
    Wong, K.-Y.K., Cipolla, R.: Reconstruction of sculpture from its profiles with unknown camera positions. IEEE Transactions on Image Processing 13(3), 381–389 (2004)CrossRefGoogle Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2006

Authors and Affiliations

  • Keith Forbes
    • 1
  • Fred Nicolls
    • 1
  • Gerhard de Jager
    • 1
  • Anthon Voigt
    • 2
  1. 1.Department of Electrical EngineeringUniversity of Cape TownSouth Africa
  2. 2.Automation and Informatics Research, De Beers Group ServicesJohannesburgSouth Africa

Personalised recommendations