A Disparity Estimation Method Based on Curvature and Belief Propagation
ZHAO Ge1,2, LIN Lan1,2, TANG Yan-Dong1, WANG Yao-Nan3
1.State Key Laboratory of Robotics, Shenyang Institute of Automation, Chinese Academy of Sciences, Shenyang 110016 2.University of Chinese Academy of Sciences, Beijing 100049 3.College of Electrical and Information Engineering, Hunan University, Changsha 410082
Abstract The current widely used matching costs are sensitive to complex optical distortions, such as vignetting. By analyzing the camera model, the robustness of the curvature-based matching cost against the vignetting effect is proved. Then, the integral image is used to compute the curvature-based matching cost so that the speed of the algorithm is improved significantly. Finally, a regularization term is designed based on curvature constraint, and thus the over-smoothing of the disparity map is avoided by restricting the belief propagation in the depth discontinuity area. Experimental results verify the effectiveness of the proposed method.
[1] Cornelis N, Leibe B, Cornelis K, et al. 3D Urban Scene Modeling Integrating Recognition and Reconstruction. International Journal of Computer Vision, 2008, 78(2/3): 121-141 [2] Micusik B, Kosecka J. Piecewise Planar City 3D Modeling from Street View Panoramic Sequence // Proc of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition. Arkansas, USA, 2009: 2906-2909 [3] Scharstein D, Sezliski R. A Taxonomy and Evaluation of Dense Two-Frame Stereo Correspondence Algorithms. International Journal of Computer Vision, 2002, 47(1/2/3): 7-42 [4] Yoon K, Kweon I. Stereo Matching with the Distinctive Similarity Measure // Proc of the IEEE Computer Society Conference on Computer Vision. Rio de Janeiro, Brazil, 2007: 1-7 [5] Einecke N, Eggert G. A Two Stage Correlation Method for Stereoscopic Depth Estimation // Proc of the International Conference on Digital Image Computing: Techniques Theory and Applications. Hong Kong, China, 2010: 227-234 [6] Evangelidis G, Psarakis E. Parametric Image Alignment Using Enhanced Correlation Coefficient Maximization. IEEE Trans on Pattern Analysis and Machine Intelligence, 2008, 30(10): 1858-1865 [7] Humenberger M, Zinner C, Michael W, et al. A Fast Stereo Ma-tching Algorithm Suitable for Real-Time Embedded System. Computer Vision and Image Understanding, 2010, 114(11): 1180-1202 [8] Humenberger M, Engelke T, Kubinger W. A Census-Based Stereo Vision Algorithm Using Modified Semi-Global Matching and Plane Fitting to Improve Matching Quality // Proc of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition. San Francisco, USA, 2010: 77-84 [9] Gu Zheng, Su Xianyu, Liu Yuankun, et al. Local Stereo Matching with Adaptive Support Weight, Rank Transform and Disparity Calibration. Pattern Recognition Letters, 2008, 29(9): 1230-1235 [10] Lahdenoja O, Alhoniemi E, Laiho M, et al. A Shape-Preserving Non-Parametric Symmetry Transform // Proc of the 18th International Conference on Pattern Recognition. Hong Kong, China, 2006: 373-377 [11] Pinheiro M, Izquierdo E, Ghanhari M, et al. Shape Matching Using a Curvature Based Polygonal Approximation in Scale Space // Proc of the IEEE International Conference on Image Processing. Genoa City, Italy, 2005, II: 538-541 [12] Saleh A, Adhami R. Curvature-Based Matching Approach for Automatic Fingerprint Identification // Proc of the 33rd Southeastern Symposium on System Theory. Singapore, Singapore, 2001: 171-175 [13] Kambhamttu C, Goldgof D. Curvature-Based Approach to Point Correspondence Recovery in Conformal Nonrigid Motion. CVGIP: Image Understanding, 1994, 60(1): 26-43 [14] Yang Q, Wang L, Yang H, et al. Stereo Matching with Color-Weighted Correlation and Hierarchical Belief Propagation and Occlusion Handling. IEEE Trans on Pattern Analysis and Machine Intelligence, 2009, 31(3): 492-504 [15] Sormann M, Konrad K, Klaus A. Segment-Based Stereo Matching Using Belief Propagation and a Self-Adapting Dissimilarity Measure // Proc of the 18th International Conference on Pattern Recognition. Hong Kong, China, 2006, III: 15-18 [16] Mitsunaga T, Nayar S. Radiometric Self Calibration // Proc of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition. New York, USA, 1999, I: 1374-1380 [17] Zhang Ke, Lu Jiangbo, Lafruit G. Cross-Based Local Stereo Ma-tching Using Orthogonal Integral Images. IEEE Trans on Circuits and Systems for Video Technology, 2009, 19(7): 1073-1079 [18] Yang Qingxiong, Yang Ruigang, Davis J, et al. Spatial-Depth Super Resolution for Range Images // Proc of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition. Minneapolis, USA, 2007: 1-8 [19] Wang Zengfu, Zheng Zhigang. A Region Based Stereo Matching Algorithm Using Cooperative Optimization // Proc of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition. Denver, USA, 2008: 1-8 [20] Orchard J, Ebrahimi M, Wong A. Efficient Nonlocal-Means Denoising Using the SVD // Proc of the IEEE International Conference on Image Processing. San Diego, USA, 2008: 1732-1735 [21] Buades A, Coll B, Morel J. Non-Local Means Denoising[EB/OL].[2012-10-10]. http://dx.doi.org/10.5201/ipol.2011.bcm_nlm
2013, Vol. 26 
