Advertisement

International Journal of Computer Vision

, Volume 59, Issue 3, pp 259–284 | Cite as

SoftPOSIT: Simultaneous Pose and Correspondence Determination

  • Philip David
  • Daniel DeMenthon
  • Ramani Duraiswami
  • Hanan Samet
Article

Abstract

The problem of pose estimation arises in many areas of computer vision, including object recognition, object tracking, site inspection and updating, and autonomous navigation when scene models are available. We present a new algorithm, called SoftPOSIT, for determining the pose of a 3D object from a single 2D image when correspondences between object points and image points are not known. The algorithm combines the iterative softassign algorithm (Gold and Rangarajan, 1996; Gold et al., 1998) for computing correspondences and the iterative POSIT algorithm (DeMenthon and Davis, 1995) for computing object pose under a full-perspective camera model. Our algorithm, unlike most previous algorithms for pose determination, does not have to hypothesize small sets of matches and then verify the remaining image points. Instead, all possible matches are treated identically throughout the search for an optimal pose. The performance of the algorithm is extensively evaluated in Monte Carlo simulations on synthetic data under a variety of levels of clutter, occlusion, and image noise. These tests show that the algorithm performs well in a variety of difficult scenarios, and empirical evidence suggests that the algorithm has an asymptotic run-time complexity that is better than previous methods by a factor of the number of image points. The algorithm is being applied to a number of practical autonomous vehicle navigation problems including the registration of 3D architectural models of a city to images, and the docking of small robots onto larger robots.

object recognition autonomous navigation POSIT softassign 

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

References

  1. Arya, S., Mount, D.M., Netanyahu, N.S., Silverman, R., and Wu, A. 1998. An optimal algorithm for approximate nearest neighbor searching. Journal of the ACM, 45(6):891–923.Google Scholar
  2. Baird, H.S. 1985. Model-Based Image Matching Using Location. MIT Press: Cambridge, MA.Google Scholar
  3. Beis, J.S. and Lowe, D.G. 1999. Indexing without invariants in 3D object recognition. IEEE Trans. Pattern Analysis and Machine Intelligence, 21(10):1000–1015.Google Scholar
  4. Beveridge, J.R. and Riseman, E.M. 1992. Hybrid weak-perspective and full-perspective matching. In Proc. IEEE Conf. Computer Vision and Pattern Recognition, Champaign, IL, pp. 432–438.Google Scholar
  5. Beveridge, J.R. and Riseman, E.M. 1995. Optimal geometric model matching under full 3D perspective. Computer Vision and Image Understanding, 61(3):351–364.Google Scholar
  6. Brand, P. and Mohr, R. 1994. Accuracy in image measure. In Proc. SPIE, Videometrics III, Boston, MA, pp. 218–228.Google Scholar
  7. Breuel, T.M. 1992. Fast recognition using adaptive subdivisions of transformation space. In Proc. IEEE Conf. on Computer Vision and Pattern Recognition, Champaign, IL, pp. 445–451.Google Scholar
  8. Bridle, J.S. 1990. Training stochastic model recognition as networks can lead to maximum mutual information estimation of parameters. In Proc. Advances in Neural Information Processing Systems, Denver, CO, pp. 211–217.Google Scholar
  9. Burns, J.B., Weiss, R.S., and Riseman, E.M. 1993. View variation of point-set and line-segment features. IEEE Trans. Pattern Analysis and Machine Intelligence, 15(1):51–68.Google Scholar
  10. Cass, T.A. 1992. Polynomial-time object recognition in the presence of clutter, occlusion, and uncertainty. In Proc. European Conf. on Computer Vision, Santa Margherita Ligure, Italy, pp. 834–842.Google Scholar
  11. Cass, T.A. 1994. Robust geometric matching for 3D object recognition. In. Proc. 12th IAPR Int. Conf. on Pattern Recognition, Jerusalem, Israel, vol. 1, pp. 477–482.Google Scholar
  12. Cass, T.A. 1998. Robust affine structure matching for 3D object recognition. IEEE Trans. on Pattern Analysis and Machine Intelligence, 20(11):1265–1274.Google Scholar
  13. DeMenthon, D. and Davis, L.S. 1993. Recognition and tracking of 3D objects by 1D search. In Proc. DARPA Image Understanding Workshop, Washington, DC, pp. 653–659.Google Scholar
  14. DeMenthon, D. and Davis, L.S. 1995. Model-based object pose in 25 lines of code. International Journal of Computer Vision, 15(1/2):123–141.Google Scholar
  15. DeMenthon, D., David, P., and Samet, H. 2001. SoftPOSIT: An algorithm for registration of 3D models to noisy perspective images combining softassign and POSIT. University of Maryland, College Park, MD, Report CS-TR-969, CS-TR 4257.Google Scholar
  16. Ely, R.W., Digirolamo, J.A., and Lundgren, J.C. 1995. Model supported positioning. In Proc. SPIE, Integrating Photogrammetric Techniques with Scene Analysis and Machine Vision II, Orlando, FL.Google Scholar
  17. Fiore, P.D. 2001. Efficient linear solution of exterior orientation. IEEE Trans. on Pattern Analysis and Machine Intelligence, 23(2):140–148.Google Scholar
  18. Fischler, M.A. and Bolles, R.C. 1981. Random sample consensus: A paradigm for model fitting with applications to image analysis and automated cartography. Comm. Association for Computing Machinery, 24(6):381–395.Google Scholar
  19. Geiger, D. and Yuille, A.L. 1991. A common framework for image segmentation. International Journal of Computer Vision, 6(3):227–243.Google Scholar
  20. Gold, S. and Rangarajan, A. 1996. Agraduated assignment algorithm for graph matching. IEEE Trans. on Pattern Analysis and Machine Intelligence, 18(4):377–388.Google Scholar
  21. Gold, S., Rangarajan, A., Lu, C.-P., Pappu, S., and Mjolsness, E. 1998. New algorithms for 2D and 3D point matching: Pose estimation and correspondence. Pattern Recognition, 31(8):1019–1031.Google Scholar
  22. Grimson, E. 1990. Object Recognition by Computer: The Role of Geometric Constraints. MIT Press: Cambridge, MA.Google Scholar
  23. Grimson, E. and Huttenlocher, D.P. 1991. On the verification of hypothesized matches in model-based recognition. IEEE Trans. on Pattern Analysis and Machine Intelligence, 13(12):1201–1213.Google Scholar
  24. Haralick, R.M., Lee, C., Ottenberg, K., and Nolle, M. 1991. Analysis and Solutions of the three point perspective pose estimation problem. In Proc. IEEE Conf. on Computer Vision and Pattern Recognition, Maui, HI, pp. 592–598.Google Scholar
  25. Hartley, R. and Zisserman, A. 2000. Multiple ViewGeometry in Computer Vision. Cambridge University Press: Cambridge, UK.Google Scholar
  26. Horaud, R., Conio, B., Leboulleux, O., and Lacolle, B. 1989. An analytic solution for the perspective 4-point problem. In Proc. IEEE Conf. on Computer Vision and Pattern Recognition, San Diego, CA, pp. 500–507.Google Scholar
  27. Horn, B.K.P. 1986. Robot Vision. MIT Press: Cambridge, MA.Google Scholar
  28. Jacobs, D.W. 1992. Space efficient 3-D model indexing. In Proc. IEEE Conf. on Computer Vision and Pattern Recognition, Champaign, IL, pp. 439–444.Google Scholar
  29. Jurie, F. 1999. Solution of the simultaneous pose and correspondence problem using gaussian error model. Computer Vision and Image Understanding, 73(3):357–373.Google Scholar
  30. Lamdan, Y. and Wolfson, H.J. 1988. Geometric hashing: A general and efficient model-based recognition scheme. In Proc. IEEE Int. Conf. on Computer Vision, Tampa, FL, pp. 238–249.Google Scholar
  31. Lu, C.-P., Hager, G.D. and Mjolsness, E. 2000. Fast and globally convergent pose estimation from video images. IEEE Trans. on Pattern Analysis and Machine Intelligence, 22(6):610–622.Google Scholar
  32. Moon, T.K. 1996. The expectation-maximization algorithm. IEEE Signal Processing Magazine, 13(6):47–60.Google Scholar
  33. Morokoff, W.J. and Caflisch, R.E., 1994. Quasi-random sequences and their discrepancies. SIAM Journal Scientific Computing, 15(6):1251–1279.Google Scholar
  34. Murase, H. and Nayar, S.K. 1995. Visual learning and recognition of 3-D objects from appearance. Int. Journal of Computer Vision, 14(1):5–24.Google Scholar
  35. Olson, C.F. 1997. Efficient pose clustering using a randomized algorithm. Int. Journal of Computer Vision, 23(2):131–147.Google Scholar
  36. Procter, S. and Illingworth, J. 1997. ForeSight: Fast object recognition using geometric hashing with edge-triple features. In Proc. Int. Conf. on Image Processing, vol. 1, Santa Barbara, CA, pp. 889–892.Google Scholar
  37. Sinkhorn, R. 1964. Arelationship between arbitrary positive matrices and doubly stochastic matrices. Annals Mathematical Statistics, 35(2):876–879.Google Scholar
  38. Ullman, S. 1989. Aligning pictorial descriptions: An approach to object recognition. Cognition, 32:193–254.Google Scholar
  39. Wunsch, P. and Hirzinger, G. 1996. Registration of CAD models to images by iterative inverse perspective matching. In Proc. Int. Conf. on Pattern Recognition, vol. 1, Vienna, Austria, pp. 78–83.Google Scholar
  40. Yuan, J.-C. 1989. A general photogrammetric method for determining object position and orientation. IEEE Trans. on Robotics and Automation, 5(2):129–142.Google Scholar

Copyright information

© Kluwer Academic Publishers 2004

Authors and Affiliations

  • Philip David
    • 1
    • 2
  • Daniel DeMenthon
    • 1
  • Ramani Duraiswami
    • 1
  • Hanan Samet
    • 1
  1. 1.University of Maryland Institute for Advanced Computer StudiesCollege ParkUSA;
  2. 2.Army Research LaboratoryAdelphi

Personalised recommendations