Accurate 3D Structure Measurements from Two Uncalibrated Views

  • Benjamin Albouy
  • Emilie Koenig
  • Sylvie Treuillet
  • Yves Lucas
Part of the Lecture Notes in Computer Science book series (LNCS, volume 4179)


We have developed an efficient algorithm to compute an Euclidean reconstruction from only two wide-baseline color images captured with a hand-held digital camera. The classical reconstruction scheme has been improved to boost the number of matches by a hierarchical epipolar constraint during an iterative process and an ultimate step of dense matching based on affine transformation. At the output, between three to four thousands points are reconstructed in 2 minutes on 1024x768 images. The stability of the algorithm has been evaluated by some repetitive tests and the quality of the reconstruction is assessed according to a metric ground truth provided by an industrial 3D scanner. The averaged error on 3D points is around 3.5% reported to the model depth. Such a precision makes this technique suitable for wound volumetric assessment in clinical environments using a hand held digital camera.


Affine Transformation Fundamental Matrix Bundle Adjustment Epipolar Line Sift Descriptor 
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.
    Scharstein, R., Szeliski, R.: A taxonomy and evaluation of dense two-frame stereo correspondance algorithms. Int. J. of Computer Vision 47(1/2/3), 7–42 (2002)MATHCrossRefGoogle Scholar
  2. 2.
    Hartley, R.I., Zisserman, A.: Multiple View geometry in Computer Vision. Cambridge University Press, Cambridge (2000)MATHGoogle Scholar
  3. 3.
    Pollefeys, M., Koch, R., Van Gool, L.: Self-calibration and metric 3D reconstruction in spite of varying and unknown internal camera parameter. In: Proc. of Int. Conf. on Computer Vision, pp. 90–95 (1998)Google Scholar
  4. 4.
    Pollefeys, M., Van Gool, L., Vergauwen, M., Verbiest, F., Cornelis, K., Tps, J.: Visual modeling with a hand-held camera. Int. J. Computer Vision 53(3), 207–232 (2004)CrossRefGoogle Scholar
  5. 5.
    Royer, E., Lhuillier, M., Dhome, M., Chateau, T.: Localization in urban environment: monocular vision compared to GPS sensor. In: Proc. of Conf. on Computer Vision and Pattern Recognition, pp. 114–121 (2005)Google Scholar
  6. 6.
    Albouy, B., Treuillet, S., Lucas, Y., Pichaud, J.C.: Volume estimation from two uncalibrated views applied to wound measurement. In: Roli, F., Vitulano, S. (eds.) ICIAP 2005. LNCS, vol. 3617, pp. 945–952. Springer, Heidelberg (2005)CrossRefGoogle Scholar
  7. 7.
    Mikolajczyk, K., Schmid, C.: Performance evaluation of local descriptors. IEEE Trans. on PAMI 27(10), 1615–1630 (2005)Google Scholar
  8. 8.
    Lowe, D.: Distinctive image feature from scale-invariant keypoints. Int. Journal of Computer Vision 2(60), 91–110 (2004)CrossRefGoogle Scholar
  9. 9.
    Strecha, C., Tuytelaars, T., Van Gool, L.: Dense matching of multiple wide-baseline views. In: IEEE Int. Conf. on Computer Vision, pp. 1194–1201 (2003)Google Scholar
  10. 10.
    Albouy, B., Treuillet, S., Lucas, Y., Birov, D.: Fundamental matrix estimation revisited through a global 3D reconstruction framework, ACIVS, Bruxels, pp. 185–192 (2004)Google Scholar
  11. 11.
    Albouy, B., Treuillet, S., Lucas, Y.: Mesure volumétrique d’escarres à partir de vues stéréoscopiques non calibrées, GRETSI, Louvain-la-Neuve (2005)Google Scholar
  12. 12.
    Sturm, P., Cheng, Z.L., Chen, P.C.Y., Poo, A.N.: Focal length calibration from two views: method and analysis of singular cases. Computer Vision and Image Understanding 99(1), 58–95 (2005)CrossRefGoogle Scholar
  13. 13.
    Gouet, V.: Mise en correpondance d’images en couleurs, application à la synthèse de vues intermédiaires, Thèse de Doctorat de l’Université de Montpellier II (2000)Google Scholar
  14. 14.
    Oram, D.: Projective Reconstruction and Metric Models from Uncalibrated Video Sequences, PhDThesis, University of Manchester (2001)Google Scholar
  15. 15.
    Lourakis, M.I.A., Argyros, A.A.: The Design and Implementation of a Generic Sparse Bundle Adjustment Software Package Based on the Levenberg-Marquardt Algorithm, Institute of Computer Science – FORTH, Heraklion, Greece , vol. 340 (2004)Google Scholar
  16. 16.
    Besl, P.J., Mc Kay, N.D.: A method for registration of 3D shape. IEEE Trans. on PAMI 14(2), 239–256 (1992)Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2006

Authors and Affiliations

  • Benjamin Albouy
    • 1
  • Emilie Koenig
    • 2
  • Sylvie Treuillet
    • 2
  • Yves Lucas
    • 3
  1. 1.ENSIVision and Robotics Laboratory, Orleans UniversityBourgesFrance
  2. 2.Polytech’Orléans, site GaliléeOrléans
  3. 3.IUT Mesures PhysiquesBourges

Personalised recommendations