Collaborative Total Variation: A General Framework for Vectorial TV Models

Even after two decades, the total variation (TV) remains one of the most popular regularizations for image processing problems and has sparked a tremendous amount of research, particularly on moving from scalar to vector-valued functions. In this paper, we consider the gradient of a color image as a three-dimensional matrix or tensor with dimensions corresponding to the spatial extent, the intensity differences between neighboring pixels, and the spectral channels. The smoothness of this tensor is then measured by taking different norms along the different dimensions. Depending on the types of these norms, one obtains very different properties of the regularization, leading to novel models for color images. We call this class of regularizations collaborative total variation (CTV). On the theoretical side, we characterize the dual norm, the subdifferential, and the proximal mapping of the proposed regularizers. We further prove, with the help of the generalized concept of singular vectors, that an $\ell^{\...

[1]  Ron Kimmel,et al.  A general framework for low level vision , 1998, IEEE Trans. Image Process..

[2]  Yehoshua Y. Zeevi,et al.  Variational denoising of partly textured images by spatially varying constraints , 2006, IEEE Transactions on Image Processing.

[3]  Tony F. Chan,et al.  High-Order Total Variation-Based Image Restoration , 2000, SIAM J. Sci. Comput..

[4]  J. Hiriart-Urruty,et al.  Convex analysis and minimization algorithms , 1993 .

[5]  Catalina Sbert,et al.  Half-linear regularization for nonconvex image restoration models , 2015 .

[6]  S. Osher,et al.  Decomposition of images by the anisotropic Rudin‐Osher‐Fatemi model , 2004 .

[7]  Michael Möller,et al.  A Novel Framework for Nonlocal Vectorial Total Variation Based on ℓ p, q, r -norms , 2015, EMMCVPR.

[8]  M. Yuan,et al.  Model selection and estimation in regression with grouped variables , 2006 .

[9]  L. Rudin,et al.  Nonlinear total variation based noise removal algorithms , 1992 .

[10]  Tony F. Chan,et al.  Color TV: total variation methods for restoration of vector-valued images , 1998, IEEE Trans. Image Process..

[11]  M. Kowalski.  Sparse regression using mixed norms , 2009 .

[12]  Yoram Singer,et al.  Efficient projections onto the l1-ball for learning in high dimensions , 2008, ICML '08.

[13]  Rob Fergus,et al.  Fast Image Deconvolution using Hyper-Laplacian Priors , 2009, NIPS.

[14]  Jahn Müller,et al.  Higher-Order TV Methods—Enhancement via Bregman Iteration , 2012, Journal of Scientific Computing.

[15]  Petros Maragos,et al.  Tensor-based image diffusions derived from generalizations of the Total Variation and Beltrami Functionals , 2010, 2010 IEEE International Conference on Image Processing.

[16]  Thomas Pock,et al.  Non-local Total Generalized Variation for Optical Flow Estimation , 2014, ECCV.

[17]  Laurent Condat,et al.  Joint demosaicking and denoising by total variation minimization , 2012, 2012 19th IEEE International Conference on Image Processing.

[18]  Martin Burger,et al.  Locally Sparse Reconstruction Using the $\ell^{1,\infty}$-Norm , 2014, 1405.5908.

[19]  Petros Maragos,et al.  Convex Generalizations of Total Variation Based on the Structure Tensor with Applications to Inverse Problems , 2013, SSVM.

[20]  Michael Möller,et al.  On the Implementation of Collaborative TV Regularization: Application to Cartoon+Texture Decomposition , 2016, Image Process. Line.

[21]  Jean-François Aujol,et al.  Projected Gradient Based Color Image Decomposition , 2009, SSVM.

[22]  Michael Möller,et al.  A Convex Model for Matrix Factorization and Dimensionality Reduction on Physical Space and Its Application to Blind Hyperspectral Unmixing , 2010 .

[23]  A. Bovik,et al.  A universal image quality index , 2002, IEEE Signal Processing Letters.

[24]  Carola-Bibiane Schönlieb,et al.  A Combined First and Second Order Variational Approach for Image Reconstruction , 2012, Journal of Mathematical Imaging and Vision.

[25]  R. Tyrrell Rockafellar,et al.  Variational Analysis , 1998, Grundlehren der mathematischen Wissenschaften.

[26]  M. Grasmair,et al.  Anisotropic Total Variation Filtering , 2010 .

[27]  Antonin Chambolle,et al.  A First-Order Primal-Dual Algorithm for Convex Problems with Applications to Imaging , 2011, Journal of Mathematical Imaging and Vision.

[28]  Tony F. Chan,et al.  Total Variation Denoising and Enhancement of Color Images Based on the CB and HSV Color Models , 2001, J. Vis. Commun. Image Represent..

[29]  Karl Kunisch,et al.  Total Generalized Variation , 2010, SIAM J. Imaging Sci..

[30]  Tony F. Chan,et al.  A General Framework for a Class of First Order Primal-Dual Algorithms for Convex Optimization in Imaging Science , 2010, SIAM J. Imaging Sci..

[31]  Laurent D. Cohen,et al.  Non-local Regularization of Inverse Problems , 2008, ECCV.

[32]  Mostafa Kaveh,et al.  Fourth-order partial differential equations for noise removal , 2000, IEEE Trans. Image Process..

[33]  Michael Möller,et al.  Color Bregman TV , 2013, SIAM J. Imaging Sci..

[34]  Mingqiang Zhu,et al.  An Efficient Primal-Dual Hybrid Gradient Algorithm For Total Variation Image Restoration , 2008 .

[35]  Kevin M. Holt,et al.  Total Nuclear Variation and Jacobian Extensions of Total Variation for Vector Fields , 2014, IEEE Transactions on Image Processing.

[36]  Joachim Weickert,et al.  Anisotropic diffusion in image processing , 1996 .

[37]  Pablo A. Parrilo,et al.  Guaranteed Minimum-Rank Solutions of Linear Matrix Equations via Nuclear Norm Minimization , 2007, SIAM Rev..

[38]  R. Tyrrell Rockafellar,et al.  Convex Analysis , 1970, Princeton Landmarks in Mathematics and Physics.

[39]  Ron Kimmel,et al.  Images as embedded maps and minimal surfaces: Movies, color,texture, and medical images , 2001 .

[40]  Silvano Di Zenzo,et al.  A note on the gradient of a multi-image , 1986, Comput. Vis. Graph. Image Process..

[41]  Tony F. Chan,et al.  Mathematical Models for Local Nontexture Inpaintings , 2002, SIAM J. Appl. Math..

[42]  Markus Grasmair,et al.  Locally Adaptive Total Variation Regularization , 2009, SSVM.

[43]  T. Chan,et al.  Fast dual minimization of the vectorial total variation norm and applications to color image processing , 2008 .

[44]  Martin Burger,et al.  Ground States and Singular Vectors of Convex Variational Regularization Methods , 2012, 1211.2057.

[45]  Guy Gilboa,et al.  Nonlocal Linear Image Regularization and Supervised Segmentation , 2007, Multiscale Model. Simul..

[46]  Yoshinori Sakai,et al.  Vectorized total variation defined by weighted L infinity norm for utilizing inter channel dependency , 2012, 2012 19th IEEE International Conference on Image Processing.

[47]  Wotao Yin,et al.  An Iterative Regularization Method for Total Variation-Based Image Restoration , 2005, Multiscale Model. Simul..

[48]  Guy Gilboa,et al.  Nonlocal Operators with Applications to Image Processing , 2008, Multiscale Model. Simul..

[49]  Petros Maragos,et al.  Structure Tensor Total Variation , 2015, SIAM J. Imaging Sci..

[50]  Guillermo Sapiro,et al.  Vector-valued active contours , 1996, Proceedings CVPR IEEE Computer Society Conference on Computer Vision and Pattern Recognition.

[51]  Daniel Cremers,et al.  The Natural Vectorial Total Variation Which Arises from Geometric Measure Theory , 2012, SIAM J. Imaging Sci..

[52]  Michael Unser,et al.  Poisson Image Reconstruction With Hessian Schatten-Norm Regularization , 2013, IEEE Transactions on Image Processing.

[53]  Xiaoming Yuan,et al.  Adaptive Primal-Dual Hybrid Gradient Methods for Saddle-Point Problems , 2013, 1305.0546.

[54]  Zhou Wei,et al.  Generalized Euler identity for subdifferentials of homogeneous functions and applications , 2008 .