Advertisement

gDLS: A Scalable Solution to the Generalized Pose and Scale Problem

  • Chris Sweeney
  • Victor Fragoso
  • Tobias Höllerer
  • Matthew Turk
Part of the Lecture Notes in Computer Science book series (LNCS, volume 8692)

Abstract

In this work, we present a scalable least-squares solution for computing a seven degree-of-freedom similarity transform. Our method utilizes the generalized camera model to compute relative rotation, translation, and scale from four or more 2D-3D correspondences. In particular, structure and motion estimations from monocular cameras lack scale without specific calibration. As such, our methods have applications in loop closure in visual odometry and registering multiple structure from motion reconstructions where scale must be recovered. We formulate the generalized pose and scale problem as a minimization of a least squares cost function and solve this minimization without iterations or initialization. Additionally, we obtain all minima of the cost function. The order of the polynomial system that we solve is independent of the number of points, allowing our overall approach to scale favorably. We evaluate our method experimentally on synthetic and real datasets and demonstrate that our methods produce higher accuracy similarity transform solutions than existing methods.

Keywords

Image Noise Similarity Transformation Polynomial System Multiple Camera Scalable Solution 
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.

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

References

  1. 1.
    Besl, P.J., McKay, N.D.: A method for registration of 3-d shapes. IEEE Transactions on Pattern Analysis and Machine Intelligence 4(2), 239–256 (1992)CrossRefGoogle Scholar
  2. 2.
    Bujnak, M., Kukelova, Z., Pajdla, T.: A general solution to the p4p problem for camera with unknown focal length. In: Proc. IEEE Conference on Computer Vision and Pattern Recognition, pp. 1–8. IEEE (2008)Google Scholar
  3. 3.
    Chen, C.S., Chang, W.Y.: On pose recovery for generalized visual sensors. IEEE Transactions on Pattern Analysis and Machine Intelligence 26(7), 848–861 (2004)CrossRefGoogle Scholar
  4. 4.
    Clemente, L.A., Davison, A.J., Reid, I.D., Neira, J., Tardós, J.D.: Mapping large loops with a single hand-held camera. In: Robotics: Science and Systems (2007)Google Scholar
  5. 5.
    Courchay, J., Dalalyan, A., Keriven, R., Sturm, P.: Exploiting loops in the graph of trifocal tensors for calibrating a network of cameras. In: Daniilidis, K., Maragos, P., Paragios, N. (eds.) ECCV 2010, Part II. LNCS, vol. 6312, pp. 85–99. Springer, Heidelberg (2010)CrossRefGoogle Scholar
  6. 6.
    Davison, A.J., Reid, I.D., Molton, N.D., Stasse, O.: Monoslam: Real-time single camera slam. IEEE Transactions on Pattern Analysis and Machine Intelligence 29(6), 1052–1067 (2007)CrossRefGoogle Scholar
  7. 7.
    Eade, E., Drummond, T.: Unified loop closing and recovery for real time monocular slam. In: Proc. British Machine Vision Conference, vol. 13, p. 136. Citeseer (2008)Google Scholar
  8. 8.
    Ess, A., Neubeck, A., Van Gool, L.J.: Generalised linear pose estimation. In: Proc. British Machine Vision Conference, pp. 1–10 (2007)Google Scholar
  9. 9.
    Fischler, M.A., Bolles, R.C.: Random sample consensus: a paradigm for model fitting with applications to image analysis and automated cartography. Communications of the ACM 24(6), 381–395 (1981)CrossRefMathSciNetGoogle Scholar
  10. 10.
    Hesch, J., Roumeliotis, S.: A direct least-squares (dls) solution for pnp. In: Proc. of the International Conference on Computer Vision. IEEE (2011)Google Scholar
  11. 11.
    Klopschitz, M., Zach, C., Irschara, A., Schmalstieg, D.: Generalized detection and merging of loop closures for video sequences. In: Proc. 3D Data Processing, Visualization, and Transmission (2008)Google Scholar
  12. 12.
    Kneip, L., Furgale, P., Siegwart, R.: Using multi-camera systems in robotics: Efficient solutions to the npnp problem. In: Proc. International Conference on Robotics and Automation, pp. 3770–3776. IEEE (2013)Google Scholar
  13. 13.
    Kneip, L., Scaramuzza, D., Siegwart, R.: A novel parametrization of the perspective-three-point problem for a direct computation of absolute camera position and orientation. In: Proc. IEEE Conference on Computer Vision and Pattern Recognition, pp. 2969–2976. IEEE (2011)Google Scholar
  14. 14.
    Kukelova, Z., Bujnak, M., Pajdla, T.: Automatic generator of minimal problem solvers. In: Forsyth, D., Torr, P., Zisserman, A. (eds.) ECCV 2008, Part III. LNCS, vol. 5304, pp. 302–315. Springer, Heidelberg (2008)CrossRefGoogle Scholar
  15. 15.
    Kukelova, Z., Bujnak, M., Pajdla, T.: Polynomial eigenvalue solutions to minimal problems in computer vision. IEEE Transactions on Pattern Analysis and Machine Intelligence 34(7), 1381–1393 (2012)CrossRefGoogle Scholar
  16. 16.
    Lepetit, V., Moreno-Noguer, F., Fua, P.: Epnp: An accurate o (n) solution to the pnp problem. International Journal of Computer Vision 81(2), 155–166 (2009)CrossRefGoogle Scholar
  17. 17.
    Li, Y., Snavely, N., Huttenlocher, D.: Location recognition using prioritized feature matching. In: Daniilidis, K., Maragos, P., Paragios, N. (eds.) ECCV 2010, Part II. LNCS, vol. 6312, pp. 791–804. Springer, Heidelberg (2010)CrossRefGoogle Scholar
  18. 18.
    Macaulay, F.: Some formulae in elimination. In: Proc. London Mathematical Society, vol. 1(1), pp. 3–27 (1902)Google Scholar
  19. 19.
    Nistér, D., Stewénius, H.: A minimal solution to the generalised 3-point pose problem. Journal of Mathematical Imaging and Vision 27(1), 67–79 (2007)CrossRefMathSciNetGoogle Scholar
  20. 20.
    Pless, R.: Using many cameras as one. In: Proc. IEEE Conference on Conference on Computer Vision and Pattern Recognition, vol. 2, pp. II–587. IEEE (2003)Google Scholar
  21. 21.
    Schweighofer, G., Pinz, A.: Globally optimal o(n) solution to the pnp problem for general camera models. In: Proc. British Machine Vision Conference, pp. 1–10 (2008)Google Scholar
  22. 22.
    Sweeney, C.: Theia Multiview Geometry Library: Tutorial & Reference. University of California, Santa Barbara, http://cs.ucsb.edu/~cmsweeney/theia
  23. 23.
    Tariq, S., Dellaert, F.: A multi-camera 6-dof pose tracker. In: Proc. International Symposium on Mixed and Augmented Reality, pp. 296–297. IEEE (2004)Google Scholar
  24. 24.
    Thrun, S., Montemerlo, M.: The graph slam algorithm with applications to large-scale mapping of urban structures. The International Journal of Robotics Research 25(5-6), 403–429 (2006)CrossRefGoogle Scholar
  25. 25.
    Umeyama, S.: Least-squares estimation of transformation parameters between two point patterns. IEEE Transactions on Pattern Analysis and Machine Intelligence 13(4), 376–380 (1991)CrossRefGoogle Scholar
  26. 26.
    Ventura, J., Arth, C., Reitmayr, G., Schmalstieg, D.: A minimal solution to the generalized pose-and-scale problem. Accepted to: IEEE Conference on Computer Vision and Pattern Recognition (2014)Google Scholar
  27. 27.
    Williams, B., Cummins, M., Neira, J., Newman, P., Reid, I., Tardós, J.: An image-to-map loop closing method for monocular slam. In: Proc. International Conference on Intelligent Robots and Systems, pp. 2053–2059. IEEE (2008)Google Scholar
  28. 28.
    Yang, J., Li, H., Jia, Y.: Go-icp: Solving 3d registration efficiently and globally optimally. In: Proc. The International Conference on Computer Vision. IEEE (2013)Google Scholar
  29. 29.
    Zheng, Y., Kuang, Y., Sugimoto, S., Astrom, K., Okutomi, M.: Revisiting the pnp problem: A fast, general and optimal solution. In: Proc. of the International Conference on Computer Vision. IEEE (December 2013)Google Scholar

Copyright information

© Springer International Publishing Switzerland 2014

Authors and Affiliations

  • Chris Sweeney
    • 1
  • Victor Fragoso
    • 1
  • Tobias Höllerer
    • 1
  • Matthew Turk
    • 1
  1. 1.University of CaliforniaSanta BarbaraUSA

Personalised recommendations