Stereo Matching Using Robust Estimators and Line Masks

Title & Authors
Stereo Matching Using Robust Estimators and Line Masks
Kim, Nak-Hyeon; Kim, Gyeong-Beom; Jeong, Seong-Jong;

Abstract
Previous area-based stereo matching algorithms find the disparity by first computing the sum of squared differences (SSD) between corresponding points using a rectangular window, and then searching the position of the minimum SSD within the disparity range. These algorithms generate relatively many matching errors around depth discontinuities, since the SSD function may fail to search for the minimum because of varying disparity profiles in such areas. In this paper, in order to improve the matching accuracy around the depth discontinuities, a new correlation function based on robust estimation technique is proposed for stereo matching. In addition, while previous stereo algorithms utilize a single rectangular window for computing the correlation function, the proposed matching algorithm utilizes 4-directional line masks additionally to reduce the matching errors further. It has been turned out that the proposed algorithm reduces matching errors around depth discontinuities significantly. Experimental results are presented in this paper, comparing the performance of the proposed technique with those of previous algorithms using both synthetic and real images.
Keywords
Corresponding;Disparity;Depth Discontinuity;Line Mask;Outlier;Projective Distortion;Robust Estimator;Stereo Matching;
Language
Korean
Cited by
References
1.
Igarashi, S., Shibukawa, K. and Kaneta, M., 1993, '3D Measurement of Shape Using Differential Stereo Vision Algorithm,' Int. J. Japan Soc. Prec. Eng., Vol. 27, No. 3, pp. 247-252

2.
Singh, S. and Digney, B., 1999, 'Autonomous Cross-Country Navigation Using Stereo Vision,' Carnegie Mellon Univ. Tech. Rep. CMU-RI-TR-99-03

3.
Wei, Guo-Qing., Braucer, W. and Hirzinger, G., 1998, 'Intensity- and Gradient-Based Stereo Matching Using Hierarchical Gaussian Basis Functions,' IEEE Trans. Pattern Anal. Machine Intell., Vol. 20, No. 11, pp. 1143-1160

4.
Grimson, W. E. L., 1985, 'Computational Experiments with a Feature Based Stereo Algorithm,' IEEE Trans. Pattern Anal. Machine Intell., pp. 17-34

5.
Kim, N. H. and Bovik, A. C., 1988, 'A Contour-Based Stereo Matching Algorithm Using Disparity Continuity,' Patt. Recog., Vol. 21, pp. 505-514

6.
Okutomi, M. and Kanade, T., 1993, 'A Multiple-Baseline Stereo,' IEEE Trans. Pattern Anal. Machine Intell., Vol. 15, No. 4, pp. 353-363

7.
Kanade, T. and Okutomi, M., 1994, 'A Stereo Matching Algorithm with an Adaptive Window: Theory and Experiment,' IEEE Trans. Pattern Anal. Machine Intell., Vol. 16, No. 9 pp. 920-923

8.
Faugeras, O., Hotz, B. and Mathieu, H., et al, 1993, 'Real Time Correlation-Based Stereo: Algorithm, Implementations and Applications,' INRIA Tech. Rep. 2013

9.
William, Todd A., 1998, 'A High-Performance Stereo Vision System for Obstacle Detection,' Carnegie Mellon Univ. Tech. Rep., CMU-RI-TR-98-24

10.
Cox, I. J., Hingornni, S. L. and Rao, S. B., 1996, 'A Maximum likelihood stereo algorithm,' Computer Vision and Image Understanding, Vol. 63, No. 3, pp.542-567

11.
Boykov, Y., Veksler, O. and Zabih, R., 1998, 'A Variable Window Approach to Early Vision,' IEEE Trans. Pattern Anal. Machine Intell., Vol. 20, No. 12, pp. 1283-1294

12.
Black, M. J. and Anandan, P., 1993, 'A Framework for the Robust Estimation of Optical Flow,' Proc. 4th IEEE Int. Conf. Computer Vision, pp. 231-236

13.
Lee, K. H. and Wohn, K. Y., 1996, 'Robust Estimation of Optical Flow Based on the Maximum Likelihood Estimators,' IEEE Trans. Inf. & Syst., Vol. 79, No. 9, pp. 1286-1295

14.
Faugeras, O., 1993, Three-Dimensional Computer Vision, MIT. Press

15.
정성종, 김경돈, 1999, '접촉식 측정프로브를 이용한 $2{\frac{1}{2}}$차원 온더머신 측정 및 검사시스템의 설계 및 해석,' 대한기계학회논문집(A), 제23권, 제1호, pp. 37-46

16.
Tsai, R. T., 1987, 'A Versatile Camera Calibration Technique for High-Accuracy 3D Machine Vision Metrology Using Off-the-shelf TV Cameras and Lens,' IEEE J. Robot. Automa., Vol. 3, pp. 323-344