Optimization Techniques for Geometric Estimation: Beyond Minimization

We overview techniques for optimal geometric estimation from noisy observations for computer vision applications. We first describe estimation techniques based on minimization of given cost functions: least squares (LS), maximum likelihood (ML), which includes reprojection error minimization (Gold Standard) as a special case, and Sampson error minimization. We then formulate estimation techniques not based on minimization of any cost function: iterative reweight, renormalization, and hyper-renormalization. Showing numerical examples, we conclude that hyper-renormalization is robust to noise and currently is the best method.

[1]  Prasanna Rangarajan,et al.  HyperLS for Parameter Estimation in Geometric Fitting , 2011, IPSJ Trans. Comput. Vis. Appl..

[2]  Kenichi Kanatani,et al.  Compact algorithm for strictly ML ellipse fitting , 2008, 2008 19th International Conference on Pattern Recognition.

[3]  金谷 健一.  Statistical optimization for geometric computation : theory and practice , 2005 .

[4]  Takayuki Okatani,et al.  On bias correction for geometric parameter estimation in computer vision , 2009, 2009 IEEE Conference on Computer Vision and Pattern Recognition.

[5]  Peter J. Rousseeuw,et al.  Robust regression and outlier detection , 1987 .

[6]  Richard I. Hartley,et al.  In Defense of the Eight-Point Algorithm , 1997, IEEE Trans. Pattern Anal. Mach. Intell..

[7]  Kenichi Kanatani,et al.  Renormalization Returns: Hyper-renormalization and Its Applications , 2012, ECCV.

[8]  Peter Meer,et al.  Heteroscedastic Regression in Computer Vision: Problems with Bilinear Constraint , 2000, International Journal of Computer Vision.

[9]  Prasanna Rangarajan,et al.  Hyper least squares fitting of circles and ellipses , 2011, Comput. Stat. Data Anal..

[10]  S. Amari,et al.  Information geometry of estimating functions in semi-parametric statistical models , 1997 .

[11]  Hirotaka Niitsuma.  HyperLS and its applications , 2011 .

[12]  Kenichi Kanatani,et al.  Ellipse Fitting with Hyperaccuracy , 2006, IEICE Trans. Inf. Syst..

[13]  B. Ripley,et al.  Robust Statistics , 2018, Encyclopedia of Mathematical Geosciences.

[14]  Wojciech Chojnacki,et al.  On the Fitting of Surfaces to Data with Covariances , 2000, IEEE Trans. Pattern Anal. Mach. Intell..

[15]  Pau Gargallo,et al.  Conic Fitting Using the Geometric Distance , 2007, ACCV.

[16]  Kenichi Kanatani,et al.  Compact Fundamental Matrix Computation , 2009, IPSJ Trans. Comput. Vis. Appl..

[17]  Takayuki Okatani,et al.  Improving accuracy of geometric parameter estimation using projected score method , 2009, 2009 IEEE 12th International Conference on Computer Vision.

[18]  Kenichi Kanatani,et al.  Unified Computation of Strict Maximum Likelihood for Geometric Fitting , 2010, Journal of Mathematical Imaging and Vision.

[19]  Jan Mayer,et al.  A numerical evaluation of preprocessing and ILU-type preconditioners for the solution of unsymmetric sparse linear systems using iterative methods , 2009, TOMS.

[20]  W. Press,et al.  Numerical Recipes in C++: The Art of Scientific Computing (2nd edn)1 Numerical Recipes Example Book (C++) (2nd edn)2 Numerical Recipes Multi-Language Code CD ROM with LINUX or UNIX Single-Screen License Revised Version3 , 2003 .

[21]  Manolis I. A. Lourakis,et al.  SBA: A software package for generic sparse bundle adjustment , 2009, TOMS.

[22]  Nikolai I. Chernov,et al.  A doubly optimal ellipse fit , 2012, Comput. Stat. Data Anal..

[23]  P. Bickel.  Efficient and Adaptive Estimation for Semiparametric Models , 1993 .

[24]  David J. Kriegman,et al.  Practical Global Optimization for Multiview Geometry , 2006, International Journal of Computer Vision.

[25]  Adrien Bartoli,et al.  Nonlinear estimation of the fundamental matrix with minimal parameters , 2004, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[26]  J. Neyman,et al.  Consistent Estimates Based on Partially Consistent Observations , 1948 .

[27]  Takayuki Okatani,et al.  Toward a statistically optimal method for estimating geometric relations from noisy data: cases of linear relations , 2003, 2003 IEEE Computer Society Conference on Computer Vision and Pattern Recognition, 2003. Proceedings..

[28]  Kenichi Kanatani.  Optimal Homography Computation with a Reliability Measure , 1998, MVA.

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

[30]  PAUL D. SAMPSON,et al.  Fitting conic sections to "very scattered" data: An iterative refinement of the bookstein algorithm , 1982, Comput. Graph. Image Process..

[31]  Kenichi Kanatani,et al.  Performance evaluation of iterative geometric fitting algorithms , 2007, Comput. Stat. Data Anal..

[32]  William H. Press,et al.  Numerical recipes in C. The art of scientific computing , 1987 .

[33]  Richard I. Hartley,et al.  Optimal Algorithms in Multiview Geometry , 2007, ACCV.

[34]  Kenichi Kanatani,et al.  Cramer-Rao Lower Bounds for Curve Fitting , 1998, Graph. Model. Image Process..

[35]  Andrew W. Fitzgibbon,et al.  Bundle Adjustment - A Modern Synthesis , 1999, Workshop on Vision Algorithms.

[36]  Nikolai I. Chernov,et al.  Statistical efficiency of curve fitting algorithms , 2003, Comput. Stat. Data Anal..

[37]  K. Kanatani,et al.  Improved algebraic methods for circle fitting , 2009 .

[38]  Wojciech Chojnacki,et al.  Rationalising the Renormalisation Method of Kanatani , 2001, Journal of Mathematical Imaging and Vision.

[39]  Richard Szeliski,et al.  Vision Algorithms: Theory and Practice , 2002, Lecture Notes in Computer Science.

[40]  Matthieu Guillaumin,et al.  Segmentation Propagation in ImageNet , 2012, ECCV.

[41]  Kenichi Kanatani,et al.  Statistical Optimization for Geometric Fitting: Theoretical Accuracy Bound and High Order Error Analysis , 2008, International Journal of Computer Vision.

[42]  Kenichi Kanatani.  Renormalization for unbiased estimation , 1993, 1993 (4th) International Conference on Computer Vision.

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

[44]  Andrew Zisserman,et al.  Multiple View Geometry in Computer Vision (2nd ed) , 2003 .

[45]  Gabriel Taubin,et al.  Estimation of Planar Curves, Surfaces, and Nonplanar Space Curves Defined by Implicit Equations with Applications to Edge and Range Image Segmentation , 1991, IEEE Trans. Pattern Anal. Mach. Intell..