SoftPOSIT: Simultaneous Pose and Correspondence Determination

  • Philip David
  • Daniel DeMenthon
  • Ramani Duraiswami
  • Hanan Samet
Conference paper
Part of the Lecture Notes in Computer Science book series (LNCS, volume 2352)


The problem of pose estimation arises in many areas of computer vision, including object recognition, object tracking, site inspection and updating, and autonomous navigation using scene models. We present a new algorithm, called SoftPOSIT, for determining the pose of a 3D object from a single 2D image in the case that correspondences between model points and image points are unknown. The algorithm combines Gold’s iterative SoftAssign algorithm [19, 20] for computing correspondences and DeMenthon’s iterative POSIT algorithm [13] for computing object pose under a full-perspective camera model. Our algorithm, unlike most previous algorithms for this problem, 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 a run-time complexity that is better than previous methods by a factor equal to the number of image points. The algorithm is being applied to the practical problem of autonomous vehicle navigation in a city through registration of a 3D architectural models of buildings to images obtained from an on-board camera.


Object recognition autonomous navigation POSIT SoftAssign 


  1. 1.
    H.S. Baird. Model-Based Image Matching Using Location, MIT Press, 1985.Google Scholar
  2. 2.
    J.S. Beis & D.G. Lowe, “Indexing Without Invariants in 3D Object Recognition,” IEEE Trans. PAMI, vol. 21, no. 10, pp. 1000–1015, 1999.Google Scholar
  3. 3.
    J.R. Beveridge & E.M. Riseman, “Optimal Geometric Model Matching Under Full 3D Perspective,” CVIU, vol. 61, pp. 351–364, May 1995.Google Scholar
  4. 4.
    J.S. Bridle. “Training Stochastic Model Recognition as Networks can Lead to Maximum Mutual Information Estimation of Parameters”, Advances in Neural Information Processing Systems, vol. 2, pp. 211–217, 1990.Google Scholar
  5. 5.
    J.B. Burns, R.S. Weiss, & E.M. Riseman, “View Variation of Point-Set and Line-Segment Features,” IEEE Trans. PAMI, vol. 15, pp. 51–68, 1993.Google Scholar
  6. 6.
    T.M. Breuel. “Fast Recognition using Adaptive Subdivisions of Transformation Space,” Proc. 1992 IEEE CVPR, pp. 445–451.Google Scholar
  7. 7.
    T.A. Cass. “Polynomial-Time Object Recognition in the Presense of Clutter, Occlusion, and Uncertainty,” Proc. ECCV’92, pp. 834–842, Springer-Verlag, 1992.Google Scholar
  8. 8.
    T.A. Cass. “Robust Geometric Matching for 3D Object Recognition,” Proc. 12th ICPR, vol. 1, pp. 477–482, 1994.Google Scholar
  9. 9.
    T.A. Cass. “Robust Affine Structure Matching for 3D Object Recognition,” IEEE Trans. PAMI, vol. 20, no. 11, pp. 1265–1274, 1998.Google Scholar
  10. 10.
    “C” code for quasi-random number generation. Available from
  11. 11.
    P. David, D. DeMenthon, R. Duraiswami, & H. Samet, “Evaluation of the SoftPOSIT Model to Image Registration Algorithm,” University of Maryland Technical Report CAR-TR-974, February 2002.Google Scholar
  12. 12.
    D. DeMenthon & L.S. Davis. “Recognition and Tracking of 3D Objects by 1D Search,” Proc. DARPA Image Understanding Workshop, Washington, DC, April 1993.Google Scholar
  13. 13.
    D. DeMenthon & L.S. Davis, “Model-Based Object Pose in 25 Lines of Code”, International Journal of Computer Vision, vol. 15, pp. 123–141, 1995.CrossRefGoogle Scholar
  14. 14.
    D. DeMenthon & P. David, “SoftPOSIT: An Algorithm for Registration of 3D Models to Noisy Perspective Images Combining SoftAssign and POSIT,” University of Maryland Technical Report CAR-TR-970, May 2001.Google Scholar
  15. 15.
    R.W. Ely, J.A. Digirolamo, & J.C. Lundgren, “Model Supported Positioning,” Integrating Photogrammetric Techniques with Scene Analysis and Machine Vision II, SPIE Aerospace Sensing and Dual Use Sensors and Controls, Orlando, April 1995.Google Scholar
  16. 16.
    P.D. Fiore. “Efficient Linear Solution of Exterior Orientation,” IEEE Trans. Pattern Analysis and Machine Intelligence, vol. 23, no. 2, pp. 140–148, February 2001.CrossRefGoogle Scholar
  17. 17.
    M.A. Fischler & R.C. Bolles, “Random Sample Consensus: A Paradigm for Model Fitting with Applications to Image Analysis and Automated Cartography,” Comm. Association for Computing Machinery, vol. 24, no. 6, pp. 381–395, June 1981.MathSciNetGoogle Scholar
  18. 18.
    D. Geiger & A.L. Yuille, “A Common Framework for Image Segmentation”, International Journal of Computer Vision, vol. 6, pp. 227–243, 1991.CrossRefGoogle Scholar
  19. 19.
    S. Gold & A. Rangarajan, “A Graduated Assignment Algorithm for Graph Matching”, IEEE Trans. PAMI, vol. 18, pp. 377–388, 1996.Google Scholar
  20. 20.
    S. Gold, A. Rangarajan, C.P. Lu, S. Pappu, and E. Mjolsness, “New Algorithms for 2D and 3D Point Matching: Pose Estimation and Correspondence”, Pattern Recognition, vol. 31, pp. 1019–1031, 1998.CrossRefGoogle Scholar
  21. 21.
    E. Grimson, Object Recognition by Computer: The Role of Geometric Constraint, MIT Press, 1990.Google Scholar
  22. 22.
    E. Grimson & D.P. Huttenlocher, “On the Verification of Hypothesized Matches in Model-Based Recognition”, IEEE Trans. PAMI, vol. 13, no. 12, pp. 1201–1213, 1991.Google Scholar
  23. 23.
    R. Hartley & A. Zisserman. Multiple View Geometry in Computer Vision. Cambridge University Press, 2000.Google Scholar
  24. 24.
    R. Horaud, B. Conio, O. Leboulleux, & B. Lacolle, “An Analytic Solution for the Perspective 4-Point Problem,” Proc. 1989 IEEE CVPR, pp. 500–507.Google Scholar
  25. 25.
    B.K.P. Horn. Robot Vision. MIT Press, Cambridge, Massachusetts 1986.Google Scholar
  26. 26.
    D.W. Jacobs. “Space Efficient 3D Model Indexing,” Proc. 1992 IEEE CVPR, pp. 439–444.Google Scholar
  27. 27.
    F. Jurie, “Solution of the Simultaneous Pose and Correspondence Problem Using Gaussian Error Model,” CVIU, vol. 73, pp. 357–373, 1999.zbMATHGoogle Scholar
  28. 28.
    Y. Lamdan & H.J. Wolfson. “Geometric Hashing: A General and Efficient Model-Based Recognition Scheme,” Proc. 1998 IEEE ICCV, pp. 238–249.Google Scholar
  29. 29.
    C.-P. Lu, G.D. Hager, & E. Mjolsness, “Fast and Globally Convergent Pose Estimation from Video Images,” IEEE Trans. PAMI, vol. 22, pp. 610–622, 2000.Google Scholar
  30. 30.
    W.J. Morokoff & R. E. Caflisch, “Quasi-Random Sequences and their Discrepancies”, SIAM J. Sci. Comput., pp. 1251–1279, 1994.Google Scholar
  31. 31.
    H. Murase & S.K. Nayar, “Visual Learning and Recognition of 3-D Objects from Appearance,” IJCV, vol. 14, pp. 5–24, 1995.CrossRefGoogle Scholar
  32. 32.
    C.F. Olson, “Efficient Pose Clustering Using a Randomized Algorithm,” IJCV, vol. 23, no. 2, pp. 131–147, 1997.CrossRefGoogle Scholar
  33. 33.
    S. Procter & J. Illingworth. “ForeSight: Fast Object Recognition using Geometric Hashing with Edge-Triple Features,” Proc. 1997 ICIP., vol. 1, pp. 889–892.Google Scholar
  34. 34.
    R. Sinkhorn. “A Relationship between Arbitrary Positive Matrices and Doubly Stochastic Matrices”, Annals Math. Statist., vol. 35, pp. 876–879, 1964.MathSciNetzbMATHCrossRefGoogle Scholar
  35. 35.
    S. Ullman, “Aligning Pictorial Descriptions: An Approach to Object Recognition,” Cognition, vol. 32, no. 3, pp. 193–254, 1989.CrossRefMathSciNetGoogle Scholar
  36. 36.
    P. Wunsch & G. Hirzinger, “Registration of CAD Models to Images by Iterative Inverse Perspective Matching”, Proc. 1996 ICPR, pp. 78–83.Google Scholar
  37. 37.
    J.-C. Yuan, “A General Photogrammetric Method for Determining Object Position and Orientation,” IEEE Trans. Robotics and Automation, vol. 5, no. 2, pp. 129–142, 1989.CrossRefGoogle Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2002

Authors and Affiliations

  • Philip David
    • 1
    • 2
  • Daniel DeMenthon
    • 3
  • Ramani Duraiswami
    • 4
  • Hanan Samet
    • 1
  1. 1.Department of Computer ScienceUniversity of MarylandCollege Park
  2. 2.Army Research LaboratoryAdelphi
  3. 3.Language and Media Processing Laboratory, Institute for Advanced Computer StudiesUniversity of MarylandCollege Park
  4. 4.Perceptual Interfaces and Reality Laboratory, Institute for Advanced Computer StudiesUniversity of MarylandCollege Park

Personalised recommendations