Finite Sample Bias of Robust Estimators in Segmentation of Closely Spaced Structures: A Comparative Study

This paper presents the design and implementation of a new comparative analytical framework for studying the usability of modern high breakdown robust estimators. The emphasis is on finding the intrinsic limits, in terms of size and relative spatial accuracy, of such techniques in solving the emerging challenges of the segmentation of fine structures. A minimum threshold for the distance between separable structures is shown to depend mainly on the scale estimation error. A scale invariant performance measure is introduced to quantify the finite sample bias of the scale estimate of a robust estimator and the measure is evaluated for some state-of-the-art high breakdown robust estimators using datasets containing at least two close but distinct structures with varying distances and inlier ratios. The results show that the new generation of density-based robust estimators (such as pbM-estimator and TSSE) have a poorer performance in problems with datasets containing only a small number of samples in each structure compared with ones based on direct processing of the residuals (such as MSSE). An important message of this paper is that an estimator that performs best in some circumstances, may not be competitive in others: particularly performance on data structures that are relatively large and/or well-separated vs closely spaced fine structures.

[1]  Rama Chellappa,et al.  Bayesian algorithms for simultaneous structure from motion estimation of multiple independently moving objects , 2005, IEEE Transactions on Image Processing.

[2]  Alireza Bab-Hadiashar,et al.  Consistency of robust estimators in multi-structural visual data segmentation , 2007, Pattern Recognit..

[3]  Yizong Cheng,et al.  Mean Shift, Mode Seeking, and Clustering , 1995, IEEE Trans. Pattern Anal. Mach. Intell..

[4]  Peter Meer,et al.  ROBUST TECHNIQUES FOR COMPUTER VISION , 2004 .

[5]  Rae-Hong Park,et al.  Robust Adaptive Segmentation of Range Images , 1998, IEEE Trans. Pattern Anal. Mach. Intell..

[6]  David Suter,et al.  Robust Optic Flow Computation , 1998, International Journal of Computer Vision.

[7]  Charles V. Stewart,et al.  Bias in robust estimation caused by discontinuities and multiple structures , 1997, IEEE Trans. Pattern Anal. Mach. Intell..

[8]  David Suter,et al.  Range and Motion Segmentation , 2000 .

[9]  Elvezio Ronchetti Robust Regression Methods and Model Selection , 2000 .

[11]  W. Gander,et al.  Adaptive Quadrature—Revisited , 2000 .

[12]  Frederick Mosteller,et al.  Exploring Data Tables, Trends and Shapes. , 1986 .

[13]  S. Shankar Sastry,et al.  Two-View Multibody Structure from Motion , 2005, International Journal of Computer Vision.

[14]  Peter J. Rousseeuw,et al.  Robust Regression and Outlier Detection , 2005, Wiley Series in Probability and Statistics.

[15]  Sing Bing Kang,et al.  Emerging Topics in Computer Vision , 2004 .

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

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

[18]  David Suter,et al.  Robust segmentation of visual data using ranked unbiased scale estimate , 1999, Robotica.

[19]  David Suter,et al.  Data Segmentation and Model Selection for Computer Vision , 1999, Springer New York.

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

[21]  Adam Krzyzak,et al.  Robust Estimation for Range Image Segmentation and Reconstruction , 1994, IEEE Trans. Pattern Anal. Mach. Intell..

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

[23]  Peter Meer,et al.  Heteroscedastic Projection Based M-Estimators , 2005, 2005 IEEE Computer Society Conference on Computer Vision and Pattern Recognition (CVPR'05) - Workshops.

[24]  Haifeng Chen,et al.  Robust regression with projection based M-estimators , 2003, Proceedings Ninth IEEE International Conference on Computer Vision.

[25]  Peter Meer,et al.  Beyond RANSAC: User Independent Robust Regression , 2006, 2006 Conference on Computer Vision and Pattern Recognition Workshop (CVPRW'06).

[26]  Dima Damen,et al.  Detecting Carried Objects in Short Video Sequences , 2008, ECCV.

[27]  Alireza Bab-Hadiashar,et al.  Range image segmentation using surface selection criterion , 2006, IEEE Transactions on Image Processing.

[28]  Dorin Comaniciu,et al.  Mean Shift: A Robust Approach Toward Feature Space Analysis , 2002, IEEE Trans. Pattern Anal. Mach. Intell..

[29]  Carlo Tomasi,et al.  Mean shift is a bound optimization , 2005, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[30]  David Suter,et al.  Robust adaptive-scale parametric model estimation for computer vision , 2004, IEEE Transactions on Pattern Analysis and Machine Intelligence.