A General Method for Feature Matching and Model Extraction

Popular algorithms for feature matching and model extraction fall into two broad categories, generate-and-test and Hough transform variations. However, both methods suffer from problems in practical implementations. Generate-and-test methods are sensitive to noise in the data. They often fail when the generated model fit is poor due to error in the selected features. Hough transform variations are somewhat less sensitive the noise, but implementations for complex problems suffer from large time and space requirements and the detection of false positives. This paper describes a general method for solving problems where a model is extracted from or fit to data that draws benefits from both generate-and-test methods and those based on the Hough transform, yielding a method superior to both. An important component of the method is the subdivision of the problem into many subproblems. This allows efficient generate-and-test techniques to be used, including the use of randomization to limit the number of subproblems that must be examined. However, the subproblems are solved using pose space analysis techniques similar to the Hough transform, which lowers the sensitivity of the method to noise. This strategy is easy to implement and results in practical algorithms that are efficient and robust. We apply this method to object recognition, geometric primitive extraction, robust regression, and motion segmentation.

[1]  Robert E. Tarjan,et al.  Applications of a planar separator theorem , 1977, 18th Annual Symposium on Foundations of Computer Science (sfcs 1977).

[2]  Erkki Oja,et al.  A new curve detection method: Randomized Hough transform (RHT) , 1990, Pattern Recognit. Lett..

[3]  Clark F. Olson.  Improving the generalized Hough transform through imperfect grouping , 1998, Image Vis. Comput..

[4]  W. Eric L. Grimson,et al.  On The Sensitivity Of The Hough Transform For Object Recognition , 1988, [1988 Proceedings] Second International Conference on Computer Vision.

[5]  James R. Bergen,et al.  A Probabilistic Algorithm for Computing Hough Transforms , 1991, J. Algorithms.

[6]  Clark F. Olson,et al.  An Approximation Algorithm for Least Median of Squares Regression , 1997, Inf. Process. Lett..

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

[8]  V. F. Leavers,et al.  Which Hough transform , 1993 .

[9]  Clark F. Olson,et al.  Constrained Hough Transforms for Curve Detection , 1999, Comput. Vis. Image Underst..

[10]  Josef Kittler,et al.  A survey of the hough transform , 1988, Comput. Vis. Graph. Image Process..

[11]  Thomas M. Breuel,et al.  Fast recognition using adaptive subdivisions of transformation space , 1992, Proceedings 1992 IEEE Computer Society Conference on Computer Vision and Pattern Recognition.

[12]  David W. Jacobs,et al.  Robust and Efficient Detection of Salient Convex Groups , 1996, IEEE Trans. Pattern Anal. Mach. Intell..

[13]  Thomas S. Huang,et al.  Motion and structure from feature correspondences: a review , 1994, Proc. IEEE.

[14]  Yonina C. Eldar,et al.  A probabilistic Hough transform , 1991, Pattern Recognit..

[15]  H. Edelsbrunner,et al.  Computing Least Median of Squares Regression Lines and Guided Topological Sweep , 1990 .

[16]  Violet F. Leavers,et al.  The dynamic generalized Hough transform: Its relationship to the probabilistic Hough transforms and an application to the concurrent detection of circles and ellipses , 1992, CVGIP Image Underst..

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

[18]  David G. Lowe,et al.  Perceptual Organization and Visual Recognition , 2012 .