Subscribe
Issue No.08 - August (2010 vol.32)
pp: 1459-1473
Romeil Sandhu , Georgia Institute of Technology, Atlanta
Allen Tannenbaum , Georgia Institute of Technology, Atlanta and Technion, Israel Institute of Technology, Haifa
ABSTRACT
In this paper, we propose a particle filtering approach for the problem of registering two point sets that differ by a rigid body transformation. Typically, registration algorithms compute the transformation parameters by maximizing a metric given an estimate of the correspondence between points across the two sets of interest. This can be viewed as a posterior estimation problem, in which the corresponding distribution can naturally be estimated using a particle filter. In this work, we treat motion as a local variation in pose parameters obtained by running a few iterations of a certain local optimizer. Employing this idea, we introduce stochastic motion dynamics to widen the narrow band of convergence often found in local optimizer approaches for registration. Thus, the novelty of our method is threefold: First, we employ a particle filtering scheme to drive the point set registration process. Second, we present a local optimizer that is motivated by the correlation measure. Third, we increase the robustness of the registration performance by introducing a dynamic model of uncertainty for the transformation parameters. In contrast with other techniques, our approach requires no annealing schedule, which results in a reduction in computational complexity (with respect to particle size) as well as maintains the temporal coherency of the state (no loss of information). Also unlike some alternative approaches for point set registration, we make no geometric assumptions on the two data sets. Experimental results are provided that demonstrate the robustness of the algorithm to initialization, noise, missing structures, and/or differing point densities in each set, on several challenging 2D and 3D registration scenarios.
INDEX TERMS
Point set registration, particle filters, pose estimation.
CITATION
Romeil Sandhu, Allen Tannenbaum, "Point Set Registration via Particle Filtering and Stochastic Dynamics", IEEE Transactions on Pattern Analysis & Machine Intelligence, vol.32, no. 8, pp. 1459-1473, August 2010, doi:10.1109/TPAMI.2009.142
REFERENCES
[1] Stanford Univ. Computer Graphics Laboratory: The Stanford 3D Scanning Repository, 1996.
[2] M. Albota, B. Aul, D. Fouche, R. Heinrichs, D. Kocher, R. Marino, J. Mooney, N. Newbury, M. O'Brien, B. Player, B. Willard, and J. Zayhowski, "Three-Dimensional Imaging Laser Radars with Geiger-Mode Avalance Photodiode Arrays," Lincoln Laboratory J., vol. 13, no. 2, pp. 351-370, 2002.
[3] P.J. Besl and N.D. McKay, "A Method for Registration of 3-D Shapes," IEEE Trans. Pattern Analysis and Machine Intelligence, vol. 14, no. 2, pp. 239-256, Feb. 1992.
[4] Y. Chen and G. Medioni, "Object Modeling by Registration of Multiple Range Images," Image Vision and Computing, vol. 10, no. 3, pp. 145-155, 1992.
[5] H. Chui and A. Rangarajan, "A New Algorithm for Non-Rigid Point Matching," Proc. IEEE Conf. Computer Vision and Pattern Recognition, vol. 2, pp. 44-51, 2000.
[6] H. Chui, A. Rangarajan, J. Zhang, and C.M. Leonard, "Unsupervised Learning of an Atlas from Unlabeled Point-Sets," IEEE Trans. Pattern Analysis and Machine Intelligence, vol. 26, no. 2, pp. 160-172, Feb. 2004.
[7] T. Dey and J. Sun, "Normal and Feature Estimations from Noisy Point Clouds," Proc. Foundations of Software Technology and Theoretical Computer Science, vol. 7, pp. 21-32, 2006.
[8] A. Doucet, N. de Freitas, and N. Gordon, Sequential Monte Carlo Methods in Practice. Springer Verlag, 2001.
[9] D.W. Eggert, A. Lorusso, and R.B. Fisher, "Estimating 3-D Rigid Body Transformations: A Comparison of Four Major Algorithms," Machine Vision and Applications, vol. 9, nos. 5/6, pp. 272-290, 1997.
[10] A.W. Fitzgibbon, "Robust Registration of 2D and 3D Point Sets," Image Vision and Computing, vol. 21, nos. 13/14, pp. 1145-1153, 2003.
[11] N. Gelfand, N.J. Mitra, L.J. Guibas, and H. Pottmann, "Robust Global Registration," Proc. Symp. Geometry Processing, vol. 255, pp. 197-207, 2005.
[12] J. Goldberger, S. Gordon, and H. Greenspan, "An Efficient Image Similarity Measure Based on Approximations of KL-Divergence between Two Gaussian Mixtures," Proc. IEEE Int'l Conf. Computer Vision, pp. 487-493, 2003.
[13] N. Gordon, D. Salmond, and A. Smith, "Novel Approach to Nonlinear/Nongaussian Bayesian State Estimation," IEE Proc. Radar and Signal Processing, vol. 140, no. 2, pp. 107-113, 1993.
[14] S. Granger and X. Pennec, "Multi-Scale EM ICP: A Fast and Robust Approach for Surface Registration," Proc. European Conf. Computer Vision, vol. 2353, pp. 418-432, 2002.
[15] B. Horn, "Closed-form Solution of Absolute Orientation Using Unit Quaternions," J. Optical Soc. Am., vol. 4, no. 1, pp. 629-634, 1987.
[16] D.F. Huber and M. Hebert, "Fully Automatic Registration of Multiple 3D Data Sets," Image Vision and Computing, vol. 21, no. 7, pp. 637-650, 2003.
[17] P.J. Huber, Robust Statistics. John Wiley & Sons, 1981.
[18] M. Isard and A. Blake, "Condensation—Conditional Density Propagation for Visual Tracking," Int'l J. Computer Vision, vol. 29, no. 1, pp. 5-28, 1998.
[19] B. Jian and B.C. Vemuri, "A Robust Algorithm for Point Set Registration Using Mixture of Gaussians," Proc. IEEE Int'l Conf. Computer Vision, vol. 2, pp. 1246-1251, 2005.
[20] A. Johnson and M. Herbert, "Using Spin Images for Efficient Object Recognition in Cluttered 3-D Scenes," IEEE Trans. Pattern Analysis and Machine Intelligence, vol. 21, no. 5, pp. 433-449, May 1999.
[21] R. Lakaemper and M. Sobel, "Correspondences between Parts of Shapes with Particle Filters," Proc. IEEE Conf. Computer Vision and Pattern Recognition , pp. 1-8, 2008.
[22] H. Li and R. Hartley, "The 3D-3D Registration Problem Revisited," Proc. IEEE Int'l Conf. Computer Vision, pp. 1-8, 2007.
[23] B. Ma and R.E. Ellis, "Surface-Based Registration with a Particle Filter," Proc. Medical Image Computing and Computer-Assisted Intervention, vol. 3216, pp. 566-573, 2004.
[24] B. Ma and R.E. Ellis, "Unified Point Selection and Surface-Based Registration Using a Particle Filter," Proc. Medical Image Computing and Computer-Assisted Intervention, vol. 3749, pp. 75-82, 2005.
[25] A. Makadia, A. PattersonIV, and K. Daniilidis, "Fully Automatic Registration of 3D Point Clouds," Proc. IEEE Conf. Computer Vision and Pattern Recognition, vol. 1, pp. 1297-1304, 2006.
[26] M. Moghari and M. Abolmaesumi, "Point-Based Rigid-Body Registration Using an Unscented Kalman Filter," IEEE Trans. Medical Imaging, vol. 26, no. 12, pp. 1708-1728, Dec. 2007.
[27] Y. Rathi, N. Vaswani, A. Tannenbaum, and A. Yezzi, "Tracking Deforming Objects Using Particle Filtering for Geometric Active Contours," IEEE Trans. Pattern Analysis and Machine Intelligence, vol. 29, no. 8, pp. 1470-1475, Aug. 2007.
[28] B. Ristic, S. Arulampalam, and N. Gordon, Beyond the Kalman Filter: Particle Filters for Tracking Applications. Artech House, 2004.
[29] R. Sandhu, S. Dambreville, and A. Tannenbaum, "Particle Filtering for Registration of 2D and 3D Point Sets with Stochastic Dynamics," Proc. IEEE Conf. Computer Vision and Pattern Recognition, pp. 1-8, 2008.
[30] M. Sofka, G. Yang, and C.V. Stewart, "Simultaneous Covariance Driven Correspondence (CDC) and Transformation Estimation in the Expectation Maximization Framework," Proc. IEEE Conf. Computer Vision and Pattern Recognition, vol. 14, pp. 1-8, 2007.
[31] C.V. Stewart, "Robust Parameter Estimation in Computer Vision," SIAM Rev., vol. 41, no. 3, pp. 513-537, 1999.
[32] Y. Tsin and T. Kanade, "A Correlation-Based Approach to Robust Point Set Registration," Proc. European Conf. Computer Vision, pp. 558-569, 2004.