Research on Fast Dense Stereo Matching Technique Using Adaptive Mask
ZHOU Jia-Li1, WU Min2, ZHOU Hua-Ping3
1.College of Science, Zhejiang University of Technology, Hangzhou 310023 2.College of Science, Zhejiang University of Science and Technology, Hangzhou 310023 3.Design Department, LG Display Nanjing Co., Ltd, Nanjing 210046
Abstract:Based on the classical phase-only correlation algorithms, a fast stereo matching method is proposed for dense point cloud. The adaptive discrete mask is used to the matching weight of the similar dense fields by estimating the gradient of matching points, thus the reconstruction precision and reliability are improved. Moreover, the proposed method also improves computational efficiency via storing and reusing the intermediate data of 2D DFT. Since the proposed algorithm satisfies SIMD model, the GPU parallel computing makes the matching process basically reach real-time . The experimental results show that the proposed fast phase-only correlation algorithm performs well in the surface reconstruction of smooth irregular diffuse objects obtained from short-baseline parallel-axis binocular stereo device, therefore it can be widely used in 3D face recognition and other fields.
[1] Goesele M, Snavely N, Curless B, et al. Multi-View Stereo for Community Photo Collections // Proc of the IEEE 11th International Conference on Computer Vision. Riode Janeiro, Brazil, 2007: 1-8 [2] Mattoccia S, Tombari F, Di Stefano L. Fast Full-Search Equivalent Template Matching by Enhanced Bounded Correlation. IEEE Trans on Image Processing, 2008, 17(4): 528-538 [3] Di Stefano L, Mattoccia S, Tombari F. ZNCC-Based Template Matching Using Bounded Partial Correlation. Pattern Recognition Letters, 2005, 26(14): 2129-2134 [4] Heo Y S, Lee K M, Lee S U. Robust Stereo Matching Using Adaptive Normalized Cross-Correlation. IEEE Trans on Pattern Analysis and Machine Intelligence, 2011, 33(4): 807-822 [5] Takita K, Aoki T, Sasaki Y, et al. High-Accuracy Subpixel Image Registration Based on Phase-Only Correlation. IEICE Trans on Fundamentals of Electronics, Communications and Computer Sciences, 2003, E86-A(8): 1925-1934 [6] Muquit M A, Shibahara T, Aoki T. A High-Accuracy Passive 3D Measurement System Using Phase-Based Image Matching. IEICE Trans on Fundamentals of Electronics, Communications and Computer Sciences, 2006, E89-A(3): 686-697 [7] Zhou Jun, Xu Yi, Yu Wurong. Phase Matching with Multiresolution Wavelet Transform // Proc of the International Society for Optical Engineering. San Jose, USA, 2002: 82-91 [8] Zhou Jiali, Zhang Shuyou, Yang Guoping. A 3D Face Reconstruction and Recognition Method Based on Passive Binocular Stereo Vision. Acta Automatica Sinica, 2009, 35(2): 123-131 (in Chinese) (周佳立,张树有,杨国平.基于双目被动立体视觉的三维人脸重构与识别.自动化学报, 2009, 35(2): 123-131) [9] Zhou Jiali, Zhang Shuyou. A Modeling and Comparison Method for Foot Based on Passive Stereo Vision. Journal of Computer-Aided Design & Computer Graphics, 2009, 21(6): 782-788 (in Chinese) (周佳立,张树有.基于双目被动立体视觉的脚型建模与比对方法.计算机辅助设计与图形学学报, 2009, 21(6): 782-788) [10] Zhou Jiali, Zhang Shuyou, Wu Min. An Improved 3D Face Reconstruction Method. Pattern Recognition and Artificial Intelligence, 2010, 23(5): 686-694 (in Chinese) (周佳立,张树有,武 敏.一种改进的三维人脸重构方法.模式识别与人工智能, 2010, 23(5): 686-694) [11] Scharstein D, Szeliski R. A Taxonomy and Evaluation of Dense Two-Frame Stereo Correspondence Algorithms. International Journal of Computer Vision, 2002, 47(1/2/3): 7-42 [12] Hirschmuller H, Innocent P R, Garibaldi J. Real-Time Correlation-Based Stereo Vision with Reduced Border Errors. International Journal of Computer Vision, 2002, 47(1/2/3): 229-246 [13] Gerrits M, Bekaert P. Local Stereo Matching with Segmentation-Based Outlier Rejection // Proc of the 3rd Canadian Conference on Computer and Robot Vision. Québec City, Canada, 2006: 66-69 [14] Yoon K J, Kweon I S. Adaptive Support-Weight Approach for Correspondence Search. IEEE Trans on Pattern Analysis and Machine Intelligence, 2006, 28(4): 650-656 [15] Zhou Jiali. A Precisely Synchronized Switching Device Using An SLR Camera as Binocular Stereo Vision: China, 201110251599. 2011-12-28 (in Chinese) (周佳立.一种单反相机作为双目的立体视觉精确同步开关装置:中国, 201110251599. 2011-12-28)