图像引导的二阶总广义变分稀疏深度图的稠密重构

doi:10.11896/j.issn.1002-137X.2015.07.067

摘要/Abstract

摘要： 利用图像颜色信息进行深度图重构,可以恢复对象边界处的深度不连续性,但无法保证对象内部的深度均匀性。为解决该问题,提出图像引导下总广义变分正则化的深度图重构模型。该模型利用扩散张量将图像提供的边缘信息引入二阶总广义变分正则项,使得重构深度在保持对象边缘的同时逼近分段仿射平面,从而保证恢复深度既保持对象边界处的不连续性,又具有对象内部的均匀性。通过Legendre-Fenchel变换将模型转换成等效的凸凹鞍点问题,从而得到高效的一阶原始对偶求解算法。实验结果表明,该方法能够恢复尖锐的对象边缘,同时保持对象内部的深度均匀性。与现有算法相比,所提方法具有更高的峰值信噪比、归一化互协方差和更低的平均绝对误差。

关键词: 深度图重构,深度不连续性,深度均匀性,总广义变分

Abstract: The depth map reconstruction using image colors may recovery the depth discontinuities at object boundaries,but will damage depth uniformities inside objects.In order to solve this problem,we formulated depth reconstruction as a convex optimization problem which is regularized by image guided total generalized variation.By incorporating image diffusion tensor into the variation regularizer,the proposed method generates piecewise smooth depth while preserving discontinuities at object boundaries.To efficiently solve the problem,a first-order primal-dual scheme was derived based on the Legendre-Fenchel transformation.Experimental results demonstrate that our method can preserve depth discontinuities at object boundaries and uniformities inside objects and outperform existing methods in terms of peak signal-to-noise ratio,normalized cross-covariance and mean absolute error.

Key words: Depth map reconstruction,Depth discontinuities,Depth uniformities,Total generalized variation

吴少群袁红星安鹏程培红. 图像引导的二阶总广义变分稀疏深度图的稠密重构[J]. 计算机科学, 2015, 42(7): 314-319. https://doi.org/10.11896/j.issn.1002-137X.2015.07.067

WU Shao-qun YUAN Hong-xing AN Peng CHENG Pei-hong. Dense Depth Map Reconstruction via Image Guided Second-order Total Generalized Variation[J]. Computer Science, 2015, 42(7): 314-319. https://doi.org/10.11896/j.issn.1002-137X.2015.07.067

参考文献

[1] 袁红星,吴少群,余辉晴,等.语义级深度迁移的2D转3D算法[J].计算机辅助设计与图形学学报,2014,26(1):72-80 Yuan Hong-xing,Wu Shao-qun,Yu Hui-qing,et al.2D-to-3D method via semantic depth transfer[J].Journal of Computer-Aided Design & Computer Graphics,2014,6(1):72-80
[2] Vosters L,Haan G D.Efficient and stable sparse-to-dense conversion for automatic 2-D to 3-D conversion[J].IEEE Transactions on Circuits and Systems for Video Technology,2013,23(3):373-386
[3] 尚斐,杜慧茜,贾云得.结合图像结构特征和近似l0范数的压缩采样恢复算法[J].计算机辅助设计与图形学学报,2010,22(11):1874-1879 Shang Fei,Du Hui-qian,Jia Yun-de.Compressive sampling image recovery with structure features and approximate l0 norm[J].Journal of Computer-Aided Design & Computer Graphics,2012,4(1):14-28
[4] 李然,干宗良,朱秀昌.基于PCA硬阈值收缩的平滑投影Landweber图像压缩感知重构[J].中国图像图形学报,2013,18(5):504-514 Li Ran,Gan Zong-liang,Zhu Xiu-chang.Smoothed projected landweber image compressed sensing reconstruction using hard thresholding based on principal components analysis[J].Journal of Image and Graphics,2013,8(5):504-514
[5] 季云云,杨震.脉冲噪声环境下高斯稀疏信源贝叶斯压缩感知重构[J].电子学报,2013,41(2):363-370 Ji Yun-yun,Yang Zhen.Bayesian compressed sensing for gaussian sparse signals in the presence of impulsive noise[J].ACTA Electronic SINICA,2013,1(2):363-370
[6] 胡文瑾,刘仲民,李战明.一种改进的小波域图像修复算法[J].计算机科学,2014,41(5):299-303 Hu Wn-jing,Liu Zhong-min,Li Zhan-ming.Improved Algorithm for Image Inpainting in Wavelet Domains[J].Computer Science,2014,1(5):299-303
[7] Dong W,Yang X,Shi G.Compressive sensing via reweightedTV and nonlocal sparsity regularization[J].Electronics Letters,2013,49(3):184-186
[8] Hu Y,Jacob M.Higher degree total variation (HDTV) regulari-zation for image recovery[J].IEEE Transactions on Image Processing,2012,21(5):2559-2571
[9] 费选,韦志辉,肖亮,等.优化加权TV的复合正则化压缩感知图像重建[J].中国图象图形学报,2014,9(2):211-218 Fei Xuan,Wei Zhi-hui,Xiao Liang,et al.Compound regularized compressed sensing image reconstructionbased on optimal reweighted TV[J].Journal of Image and Graphics,2014,9(2):211-218
[10] Zhang J,Liu S H,Xiong R Q,et al.Improved total variation based image compressive sensing recovery by nonlocal regularization[C]∥Proc of IEEE International Symposium on Circuits and Systems.Los Alamitos:IEEE Computer Society Press,2013:2836-2839
[11] 袁红星,吴少群,朱仁祥,等.融合对象性和视觉显著度的单目图像2D转3D[J].中国图象图形学报,2013,8(11):1478-1485 Yuan Hong-xing,Wu Shao-qun,Zhu Ren-xiang,et al.Single-view image 2D-to-3D conversion based on abjectness and visual saliency[J].Journal of Image and Graphics,2013,8(11):1478-1485
[12] Hawe S,Kleinsteuber M,Diepold K.Dense disparity maps from sparse disparity measurements[C]∥Proc of IEEE International Conference on Computer Vision.Los Alamitos:IEEE Computer Society Press,2011:2126-2133
[13] Levin A,Lischinski D,Weiss Y.A closed-form solution to natural image matting[J].IEEE Transactions on Pattern Analysis and Machine Intelligence,2008,30(2):228-242
[14] Zhuo S J,Sim T.Defocus map estimation from a single image[J].Pattern Recognition,2011,44(9):1852-1858
[15] Vosters L,Haan G D.Efficient and stable sparse-to-dense conversion for automatic 2-D to 3-D conversion.IEEE Transactions on Circuits and Systems for Video Technology,2013,23(3):373-386
[16] Saad Y.Iterative methods for sparse linear systems(2nd ed)[M].Philadelphia,PA:Society for Industrial and Applied Mathe-matics,2003:1-10
[17] Wu H,Song Z,Yao J,et al.Stereo matching based on support points propagation[C]∥Proc of IEEE International Conference on Information Science and Technology.Los Alamitos:IEEE Computer Society Press,2012:732-736
[18] Huang J,Schonfeld D.A novel particle filtering framework for 2D-to-3D conversion from a monoscopic 2D image sequence[C]∥Proc of IEEE Conference on Visual Communications and Image Processing.Los Alamitos:IEEE Computer Society Press,2012:1-6
[19] Ranftl R,Gehrig S,Pock T,et al.Pushing the limits of stereousing variational stereo estimation[C]∥Proc of IEEE Intelligent Vehicles Symposium.Los Alamitos:IEEE Computer Society Press,2012:401-407
[20] Ferstl D,Reinbacher C,Ranftl R.Image guided depth upsam-pling using anisotropic total generalized variation//Proc of IEEE International Conference on Computer Vision.Los Alamitos:IEEE Computer Society Press,2013:1-8
[21] Rockafellar R T.Convex analysis[M].Princeton,NJ:Princeton University Press,1997:1-20
[22] Bredies K,Kunisch K,Pock T.Total generalized variation[J].SIAM Journal on Imaging Sciences,2010,3(3):492-526
[23] Chambolle A,Pock T.A first-order primal-dual algorithm forconvex problems with applications to imaging[J].Journal of Mathematical Imaging and Vision,2011,40(1):120-145
[24] Boyd S,Vandenberghe L.Convex optimization[M].Cambridge:Cambridge University Press,2004:1-10
[25] Pock T,Chambolle A.Diagonal preconditioning for first order primal-dual algorithms in convex optimization[C]∥Proc of IEEE International Conference on Computer Vision.Los Alamitos:IEEE Computer Society Press,2011:1762-1769
[26] Cao X,Li Z,Dai Q H.Semi-automatic 2D-to-3D conversionusing disparity propagation[J].IEEE Transactions on Broadcas-ting,2011,57(2):491-499

Metrics

Viewed

Full text

Abstract

Cited

Shared

Discussed