Monte Carlo extrinsic estimators of manifold-valued parameters

Monte Carlo (MC) methods have become an important tool for inferences in non-Gaussian and non-Euclidean settings. We study their use in those signal/image processing scenarios where the parameter spaces are certain Riemannian manifolds (finite-dimensional Lie groups and their quotient sets). We investigate the estimation of means and variances of the manifold-valued parameters, using two popular sampling methods: independent and importance sampling. Using Euclidean embeddings, we specify a notion of extrinsic means, employ Monte Carlo methods to estimate these means, and utilize large-sample asymptotics to approximate the estimator covariances. Experimental results are presented for target pose estimation (orthogonal groups) and signal subspace estimation (Grassmann manifolds). Asymptotic covariances are utilized to construct confidence regions, to compare estimators, and to determine the sample size for MC sampling.

[1]  W. Boothby.  An introduction to differentiable manifolds and Riemannian geometry , 1975 .

[2]  Richard S. Bucy,et al.  Geometry and multiple direction estimation , 1991, Inf. Sci..

[3]  Timothy J. Robinson,et al.  Sequential Monte Carlo Methods in Practice , 2003 .

[4]  R. Bhattacharya,et al.  Nonparametic estimation of location and dispersion on Riemannian manifolds , 2002 .

[5]  D. Kendall.  SHAPE MANIFOLDS, PROCRUSTEAN METRICS, AND COMPLEX PROJECTIVE SPACES , 1984 .

[6]  K. Mardia.  Statistics of Directional Data , 1972 .

[7]  U. Grenander,et al.  Computational anatomy: an emerging discipline , 1998 .

[8]  Anuj Srivastava,et al.  A Bayesian approach to geometric subspace estimation , 2000, IEEE Trans. Signal Process..

[9]  Anuj Srivastava,et al.  Bayesian filtering for tracking pose and location of rigid targets , 2000, SPIE Defense + Commercial Sensing.

[10]  K. Kanatani,et al.  Statistical optimization and geometric inference in computer vision , 1998, Philosophical Transactions of the Royal Society of London. Series A: Mathematical, Physical and Engineering Sciences.

[11]  C. Goodall.  Procrustes methods in the statistical analysis of shape , 1991 .

[12]  Alan Edelman,et al.  The Geometry of Algorithms with Orthogonality Constraints , 1998, SIAM J. Matrix Anal. Appl..

[13]  Hoon Kim,et al.  Monte Carlo Statistical Methods , 2000, Technometrics.

[14]  G. S. Watson.  Statistics on Spheres , 1983 .

[15]  David G. Kendall,et al.  Shape & Shape Theory , 1999 .

[16]  G. Stiny.  Shape , 1999 .

[17]  K. Mardia,et al.  ‘Shape, Procrustes tangent projections and bilateral symmetry’ , 2001 .

[18]  H. Hendriks.  A Crame´r-Rao–type lower bound for estimators with values in a manifold , 1991 .

[19]  Anuj Srivastava,et al.  Geometric Filtering for Subspace Tracking , 2000 .

[20]  A. V. D. Vaart,et al.  Asymptotic Statistics: Frontmatter , 1998 .

[21]  金谷 健一.  Group-theoretical methods in image understanding , 1990 .

[22]  H. C. Longuet-Higgins,et al.  A computer algorithm for reconstructing a scene from two projections , 1981, Nature.

[23]  Michael I. Miller,et al.  Hilbert-Schmidt Lower Bounds for Estimators on Matrix Lie Groups for ATR , 1998, IEEE Trans. Pattern Anal. Mach. Intell..

[24]  Olivier D. Faugeras,et al.  Characterizing the Uncertainty of the Fundamental Matrix , 1997, Comput. Vis. Image Underst..

[25]  T. K. Carne,et al.  Shape and Shape Theory , 1999 .

[26]  J. M. Oller.  On an intrinsic analysis of statistical estimation , 1993 .