Heteroscedastic Regression in Computer Vision: Problems with Bilinear Constraint

We present an algorithm to estimate the parameters of a linear model in the presence of heteroscedastic noise, i.e., each data point having a different covariance matrix. The algorithm is motivated by the recovery of bilinear forms, one of the fundamental problems in computer vision which appears whenever the epipolar constraint is imposed, or a conic is fit to noisy data points. We employ the errors-in-variables (EIV) model and show why already at moderate noise levels most available methods fail to provide a satisfactory solution. The improved behavior of the new algorithm is due to two factors: taking into account the heteroscedastic nature of the errors arising from the linearization of the bilinear form, and the use of generalized singular value decomposition (GSVD) in the computations. The performance of the algorithm is compared with several methods proposed in the literature for ellipse fitting and estimation of the fundamental matrix. It is shown that the algorithm achieves the accuracy of nonlinear optimization techniques at much less computational cost.

[1]  Olivier D. Faugeras,et al.  The fundamental matrix: Theory, algorithms, and stability analysis , 2004, International Journal of Computer Vision.

[2]  Azriel Rosenfeld,et al.  Robust regression methods for computer vision: A review , 1991, International Journal of Computer Vision.

[3]  Kenichi Kanatani,et al.  Unbiased Estimation and Statistical Analysis of 3-D Rigid Motion from Two Views , 1993, IEEE Trans. Pattern Anal. Mach. Intell..

[4]  Yoram Leedan.  Statistical analysis of quadratic problems in computer vision , 1997 .

[5]  Andrea J. van Doorn,et al.  The Generic Bilinear Calibration-Estimation Problem , 2004, International Journal of Computer Vision.

[6]  Wayne A. Fuller,et al.  Measurement Error Models , 1988 .

[7]  Kenichi Kanatani,et al.  Statistical Bias of Conic Fitting and Renormalization , 1994, IEEE Trans. Pattern Anal. Mach. Intell..

[8]  G. G. Stokes.  "J." , 1890, The New Yale Book of Quotations.

[9]  Takeo Kanade,et al.  A unified factorization algorithm for points, line segments and planes with uncertainty models , 1998, Sixth International Conference on Computer Vision (IEEE Cat. No.98CH36271).

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

[11]  J. Mendel.  Lessons in Estimation Theory for Signal Processing, Communications, and Control , 1995 .

[12]  Richard Hartley,et al.  Minimizing algebraic error , 1998, Philosophical Transactions of the Royal Society of London. Series A: Mathematical, Physical and Engineering Sciences.

[14]  O. Faugeras.  Three-dimensional computer vision: a geometric viewpoint , 1993 .

[15]  Zhengyou Zhang,et al.  Determining the Epipolar Geometry and its Uncertainty: A Review , 1998, International Journal of Computer Vision.

[16]  Andrew W. Fitzgibbon,et al.  Direct least squares fitting of ellipses , 1996, Proceedings of 13th International Conference on Pattern Recognition.

[17]  J. Vandewalle,et al.  Analysis and properties of the generalized total least squares problem AX≈B when some or all columns in A are subject to error , 1989 .

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

[19]  Takeo Kanade,et al.  Shape and motion from image streams under orthography: a factorization method , 1992, International Journal of Computer Vision.

[20]  Andrew W. Fitzgibbon,et al.  Direct Least Square Fitting of Ellipses , 1999, IEEE Trans. Pattern Anal. Mach. Intell..

[21]  Wojciech Chojnacki,et al.  Fitting surfaces to data with covariance information: Fundamental methods applicable to computer vision , 1999 .

[22]  Zhengyou Zhang,et al.  On the Optimization Criteria Used in Two-View Motion Analysis , 1998, IEEE Trans. Pattern Anal. Mach. Intell..

[23]  Joshua B. Tenenbaum,et al.  Learning bilinear models for two-factor problems in vision , 1997, Proceedings of IEEE Computer Society Conference on Computer Vision and Pattern Recognition.

[24]  Gabriel Taubin,et al.  An improved algorithm for algebraic curve and surface fitting , 1993, 1993 (4th) International Conference on Computer Vision.

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

[26]  Roger Mohr,et al.  Epipole and fundamental matrix estimation using virtual parallax , 1995, Proceedings of IEEE International Conference on Computer Vision.

[27]  Yakup Genc,et al.  Epipolar Geometry and Linear Subspace Methods: A New Approach to Weak Calibration , 2004, International Journal of Computer Vision.

[28]  Philip H. S. Torr,et al.  The Development and Comparison of Robust Methods for Estimating the Fundamental Matrix , 1997, International Journal of Computer Vision.

[29]  TaubinGabriel.  Estimation of Planar Curves, Surfaces, and Nonplanar Space Curves Defined by Implicit Equations with Applications to Edge and Range Image Segmentation , 1991 .

[30]  Rachid Deriche,et al.  A Robust Technique for Matching two Uncalibrated Images Through the Recovery of the Unknown Epipolar Geometry , 1995, Artif. Intell..

[31]  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..

[32]  Peter Meer,et al.  Optimal rigid motion estimation and performance evaluation with bootstrap , 1999, Proceedings. 1999 IEEE Computer Society Conference on Computer Vision and Pattern Recognition (Cat. No PR00149).

[33]  L. Mirsky,et al.  An introduction to linear algebra , 1957, Mathematical Gazette.

[34]  Peter Meer,et al.  Unbiased Estimation of Ellipses by Bootstrapping , 1996, IEEE Trans. Pattern Anal. Mach. Intell..

[35]  Olivier Faugeras,et al.  Three-Dimensional Computer Vision , 1993 .

[36]  Zhengyou Zhang,et al.  Parameter estimation techniques: a tutorial with application to conic fitting , 1997, Image Vis. Comput..

[37]  Sabine Van Huffel,et al.  The total least squares problem , 1993 .

[38]  O. Faugeras.  Stratification of three-dimensional vision: projective, affine, and metric representations , 1995 .

[39]  Gene H. Golub,et al.  Matrix computations , 1983 .