Conjugate gradient on Grassmann manifolds for robust subspace estimation

Most geometric computer vision problems involve orthogonality constraints. An important subclass of these problems is subspace estimation, which can be equivalently formulated into an optimization problem on Grassmann manifolds. In this paper, we propose to use the conjugate gradient algorithm on Grassmann manifolds for robust subspace estimation in conjunction with the recently introduced generalized projection based M-Estimator (gpbM). The gpbM method is an elemental subset-based robust estimation algorithm that can process heteroscedastic data without any user intervention. We show that by optimizing the orthogonal parameter matrix on Grassmann manifolds, the performance of the gpbM algorithm improves significantly. Results on synthetic and real data are presented.

[1]  Yakup Genc,et al.  Robust unambiguous parametrization of the essential manifold , 2008, 2008 IEEE Conference on Computer Vision and Pattern Recognition.

[2]  Ying Wu,et al.  Multibody Grouping by Inference of Multiple Subspaces from High-Dimensional Data Using Oriented-Frames , 2006, IEEE Trans. Pattern Anal. Mach. Intell..

[3]  Shuicheng Yan,et al.  Pursuing Informative Projection on Grassmann Manifold , 2006, 2006 IEEE Computer Society Conference on Computer Vision and Pattern Recognition (CVPR'06).

[4]  Marc Pollefeys,et al.  A General Framework for Motion Segmentation: Independent, Articulated, Rigid, Non-rigid, Degenerate and Non-degenerate , 2006, ECCV.

[5]  Fatih Murat Porikli,et al.  Learning on lie groups for invariant detection and tracking , 2008, 2008 IEEE Conference on Computer Vision and Pattern Recognition.

[6]  Peter Meer,et al.  Estimation of Nonlinear Errors-in-Variables Models for Computer Vision Applications , 2006, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[7]  Peter Meer,et al.  Simultaneous multiple 3D motion estimation via mode finding on Lie groups , 2005, Tenth IEEE International Conference on Computer Vision (ICCV'05) Volume 1.

[8]  A. Ruszczynski,et al.  Nonlinear Optimization , 2006 .

[9]  Y. Chikuse.  Statistics on special manifolds , 2003 .

[10]  P. Thomas Fletcher,et al.  Statistics of shape via principal geodesic analysis on Lie groups , 2003, 2003 IEEE Computer Society Conference on Computer Vision and Pattern Recognition, 2003. Proceedings..

[11]  Nicholas Ayache,et al.  Geometric Means in a Novel Vector Space Structure on Symmetric Positive-Definite Matrices , 2007, SIAM J. Matrix Anal. Appl..

[12]  K.A. Gallivan,et al.  Efficient algorithms for inferences on Grassmann manifolds , 2004, IEEE Workshop on Statistical Signal Processing, 2003.

[13]  Peter Meer,et al.  Nonlinear Mean Shift over Riemannian Manifolds , 2009, International Journal of Computer Vision.

[14]  Anuj Srivastava,et al.  Bayesian and geometric subspace tracking , 2004, Advances in Applied Probability.

[15]  Richard I. Hartley,et al.  Reconstruction from Projections Using Grassmann Tensors , 2004, International Journal of Computer Vision.

[16]  P. Perona,et al.  Motion estimation via dynamic vision , 1994, Proceedings of 1994 33rd IEEE Conference on Decision and Control.

[17]  Xiaoqin Zhang,et al.  Visual tracking via incremental Log-Euclidean Riemannian subspace learning , 2008, 2008 IEEE Conference on Computer Vision and Pattern Recognition.

[18]  Xavier Pennec,et al.  A Riemannian Framework for Tensor Computing , 2005, International Journal of Computer Vision.

[19]  Jan-Michael Frahm,et al.  A Comparative Analysis of RANSAC Techniques Leading to Adaptive Real-Time Random Sample Consensus , 2008, ECCV.

[20]  Fatih Murat Porikli,et al.  Pedestrian Detection via Classification on Riemannian Manifolds , 2008, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[21]  Maher Moakher,et al.  To appear in: SIAM J. MATRIX ANAL. APPL. MEANS AND AVERAGING IN THE GROUP OF ROTATIONS∗ , 2002 .

[22]  Tieniu Tan,et al.  Visual tracking via incremental self-tuning particle filtering on the affine group , 2010, 2010 IEEE Computer Society Conference on Computer Vision and Pattern Recognition.

[23]  Vittorio Murino,et al.  Multi-class Classification on Riemannian Manifolds for Video Surveillance , 2010, ECCV.

[24]  Peter Meer,et al.  Generalized projection based M-estimator: Theory and applications , 2011, CVPR 2011.

[25]  Yuri Ivanov,et al.  Fast Approximate Nearest Neighbor Methods for Non-Euclidean Manifolds with Applications to Human Activity Analysis in Videos , 2010, ECCV.

[26]  J. Shewchuk.  An Introduction to the Conjugate Gradient Method Without the Agonizing Pain , 1994 .

[27]  H. Karcher.  Riemannian center of mass and mollifier smoothing , 1977 .

[28]  William H. Press,et al.  Numerical recipes in C , 2002 .

[29]  John B. Moore,et al.  Essential Matrix Estimation Using Gauss-Newton Iterations on a Manifold , 2007, International Journal of Computer Vision.

[30]  Venu Madhav Govindu.  Lie-algebraic averaging for globally consistent motion estimation , 2004, CVPR 2004.

[31]  P. Perona,et al.  Recursive Motion Estimation on the Essential Manifold , 1993 .

[32]  Rama Chellappa,et al.  Statistical Computations on Grassmann and Stiefel Manifolds for Image and Video-Based Recognition , 2011, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[33]  Hasan Ertan Ceting.  Intrinsic Mean Shift for Clustering on Stiefel and Grassmann Manifolds , 2009 .

[34]  Robert C. Bolles,et al.  Random sample consensus: a paradigm for model fitting with applications to image analysis and automated cartography , 1981, CACM.

[35]  S. Shankar Sastry,et al.  Optimization Criteria and Geometric Algorithms for Motion and Structure Estimation , 2001, International Journal of Computer Vision.

[36]  William H. Press,et al.  Numerical recipes , 1990 .

[37]  P. Thomas Fletcher,et al.  Statistics of Shape via Principal Component Analysis on Lie Groups , 2003 .

[38]  Matthijs C. Dorst.  Distinctive Image Features from Scale-Invariant Keypoints , 2011 .

[39]  W. Eric L. Grimson,et al.  Learning visual flows: A Lie algebraic approach , 2009, CVPR.

[40]  J. Ross Beveridge,et al.  Action classification on product manifolds , 2010, 2010 IEEE Computer Society Conference on Computer Vision and Pattern Recognition.

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

[42]  Peter Meer,et al.  Subspace Estimation Using Projection Based M-Estimators over Grassmann Manifolds , 2006, ECCV.

[43]  B. O'neill.  Semi-Riemannian Geometry With Applications to Relativity , 1983 .