Convex Relaxation of Vectorial Problems with Coupled Regularization

We propose convex relaxations for nonconvex energies on vector-valued functions which are tractable yet as tight as possible. In contrast to existing relaxations, we can handle the combination of nonconvex data terms with coupled regularizers such as $l^2$-regularizers. The key idea is to consider a collection of hypersurfaces with a relaxation that takes into account the entire functional rather than separately treating the data term and the regularizers. We provide a theoretical analysis, detail the implementations for different functionals, present run time and memory requirements, and experimentally demonstrate that the coupled $l^2$-regularizers give systematic improvements regarding denoising, inpainting, and optical flow estimation.

[1]  Daniel Cremers,et al.  The Natural Total Variation Which Arises from Geometric Measure Theory , 2012 .

[2]  Christoph Schnörr,et al.  Continuous Multiclass Labeling Approaches and Algorithms , 2011, SIAM J. Imaging Sci..

[3]  C. Villani.  Topics in Optimal Transportation , 2003 .

[4]  D. Schlesinger,et al.  TRANSFORMING AN ARBITRARY MINSUM PROBLEM INTO A BINARY ONE , 2006 .

[5]  H. Fédérer,et al.  Real Flat Chains, Cochains and Variational Problems , 1974 .

[6]  G. Bouchitte,et al.  Multifonctions s.c.i. et régularisée s.c.i. essentielle , 1989 .

[7]  Daniel Cremers,et al.  Multiview Stereo and Silhouette Consistency via Convex Functionals over Convex Domains , 2011, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[8]  Daniel Cremers,et al.  A convex approach for computing minimal partitions , 2008 .

[9]  Olga Veksler,et al.  Fast approximate energy minimization via graph cuts , 2001, Proceedings of the Seventh IEEE International Conference on Computer Vision.

[10]  Martin J. Wainwright,et al.  MAP estimation via agreement on trees: message-passing and linear programming , 2005, IEEE Transactions on Information Theory.

[11]  Daniel Cremers,et al.  Global Solutions of Variational Models with Convex Regularization , 2010, SIAM J. Imaging Sci..

[12]  Daniel Cremers,et al.  Tight Convex Relaxations for Vector-Valued Labeling , 2013, SIAM J. Imaging Sci..

[13]  G. Bouchitté,et al.  The calibration method for the Mumford-Shah functional and free-discontinuity problems , 2001, math/0105013.

[14]  Vladimir Kolmogorov,et al.  What energy functions can be minimized via graph cuts? , 2002, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[15]  Christoph Schnörr,et al.  Convex optimization for multi-class image labeling with a novel family of total variation based regularizers , 2009, 2009 IEEE 12th International Conference on Computer Vision.

[16]  Marc Pollefeys,et al.  What is optimized in tight convex relaxations for multi-label problems? , 2012, 2012 IEEE Conference on Computer Vision and Pattern Recognition.

[17]  L. Mirsky.  SYMMETRIC GAUGE FUNCTIONS AND UNITARILY INVARIANT NORMS , 1960 .

[18]  Daniel Cremers,et al.  A Convex Approach to Minimal Partitions , 2012, SIAM J. Imaging Sci..

[19]  Mila Nikolova,et al.  Algorithms for Finding Global Minimizers of Image Segmentation and Denoising Models , 2006, SIAM J. Appl. Math..

[20]  Daniel Cremers,et al.  Tight convex relaxations for vector-valued labeling problems , 2011, 2011 International Conference on Computer Vision.

[21]  Antonin Chambolle,et al.  Diagonal preconditioning for first order primal-dual algorithms in convex optimization , 2011, 2011 International Conference on Computer Vision.

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

[23]  Xue-Cheng Tai,et al.  Global Minimization for Continuous Multiphase Partitioning Problems Using a Dual Approach , 2011, International Journal of Computer Vision.

[24]  L. Ambrosio,et al.  Functions of Bounded Variation and Free Discontinuity Problems , 2000 .

[25]  J. P. McKelvey,et al.  Simple transcendental expressions for the roots of cubic equations , 1984 .

[26]  L. Evans.  Measure theory and fine properties of functions , 1992 .

[27]  M. Giaquinta.  Cartesian currents in the calculus of variations , 1983 .

[28]  Ronald F. Gariepy.  FUNCTIONS OF BOUNDED VARIATION AND FREE DISCONTINUITY PROBLEMS (Oxford Mathematical Monographs) , 2001 .

[29]  Guy Bouchitté,et al.  Integral representation of convex functionals on a space of measures , 1988 .

[30]  T. O’Neil.  Geometric Measure Theory , 2002 .

[31]  Yasuyuki Matsushita,et al.  Motion detail preserving optical flow estimation , 2010, 2010 IEEE Computer Society Conference on Computer Vision and Pattern Recognition.

[32]  D. Greig,et al.  Exact Maximum A Posteriori Estimation for Binary Images , 1989 .

[33]  Michael J. Black,et al.  Secrets of optical flow estimation and their principles , 2010, 2010 IEEE Computer Society Conference on Computer Vision and Pattern Recognition.

[34]  Vladimir Kolmogorov,et al.  Computing geodesics and minimal surfaces via graph cuts , 2003, Proceedings Ninth IEEE International Conference on Computer Vision.

[35]  Davi Geiger,et al.  Segmentation by grouping junctions , 1998, Proceedings. 1998 IEEE Computer Society Conference on Computer Vision and Pattern Recognition (Cat. No.98CB36231).

[36]  Richard Szeliski,et al.  A Database and Evaluation Methodology for Optical Flow , 2007, 2007 IEEE 11th International Conference on Computer Vision.

[37]  Gianni Dal Maso,et al.  Integral representation on BV(ω) of Γ-limits of variational integrals , 1979 .

[38]  Daniel Cremers,et al.  A Convex Formulation of Continuous Multi-label Problems , 2008, ECCV.

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

[40]  Hiroshi Ishikawa,et al.  Exact Optimization for Markov Random Fields with Convex Priors , 2003, IEEE Trans. Pattern Anal. Mach. Intell..

[41]  丸山 徹.  Convex Analysisの二,三の進展について , 1977 .

[42]  Jan-Michael Frahm,et al.  Fast Global Labeling for Real-Time Stereo Using Multiple Plane Sweeps , 2008, VMV.

[43]  Thomas Brox,et al.  High Accuracy Optical Flow Estimation Based on a Theory for Warping , 2004, ECCV.

[44]  J. Sivaloganathan.  CARTESIAN CURRENTS IN THE CALCULUS OF VARIATIONS I: CARTESIAN CURRENTS (Ergebnisse der Mathematik und ihrer Grenzgebiete (3), A Series of Modern Surveys in Mathematics 37) , 2000 .