Abstract
Variational approaches are an established paradigm in the field of image processing . The non-convexity of the functional can be addressed by functional lifting and convex relaxation techniques, which aim to solve a convex approximation of the original energy on a larger space. However, so far these approaches have been limited to first-order , gradient-based regularizers such as the total variation . In this work, we propose a way to extend functional lifting to a second-order regularizer derived from the Laplacian. We prove that it can be represented efficiently and thus allows numerical optimization. We experimentally demonstrate the usefulness on a synthetic convex denoising problem and on synthetic as well as real-world image registration problems.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Notes
- 1.
See [28] and http://github.com/tum-vision/prost for the most recent version.
References
L. Ambrosio, N. Fusco, D. Pallara, Functions of Bounded Variation and Free Discontinuity Problems (Clarendon Press, Oxford, 2000)
K. Bredies, K. Kunisch, T. Pock, Total generalized variation. J. Imag. Sci. 3(3), 294–526 (2010)
T. Brox, A. Bruhn, N. Papenberg, J. Weickert, High accuracy optical flow estimation based on a theory for warping, in European Conference on Computer Vision (Springer, Berlin, 2004), pp. 25–36
A. Chambolle, An algorithm for total variation minimization and applications. J. Math. Imag. Vis. 20, 89–97 (2004)
A. Chambolle, T. Pock, A first-order primal-dual algorithm for convex problems with applications to imaging. J. Math. Imag. Vis. 40(1), 120–145 (2011)
A. Chambolle, D. Cremers, T. Pock, A convex approach to minimal partitions. J. Imag. Sci. 5(4), 1113–1158 (2012)
T. Chan, A. Marquina, P. Mulet, Higher-order total variation-based image restoration. J. Sci. Comput. 22(2), 503–516 (2000)
T. Chan, S. Esedoḡlu, F. Park, A. Yip, Total variation image restoration: overview and recent developments, in The Handbook of Mathematical Models in Computer Vision (Springer, Berlin, 2005)
F. Demengel, Fonctions à Hessien borné. Ann. Inst. Fourier 34, 155–190 (1985)
B. Fischer, J. Modersitzki, Curvature based image registration. J. Math. Imag. Vis. 18(1), 81–85 (2003)
W.J. Fu, Penalized regressions: the bridge versus the lasso. J. Comput. Graph. Stat. 7(3), 397–416 (1998)
M. Grant, S. Boyd, Graph implementations for nonsmooth convex programs, in Recent Advances in Learning and Control (Springer, Berlin, 2008), pp. 95–110
M. Grant, S. Boyd, CVX: matlab software for disciplined convex programming (2014). http://cvxr.com/cvx
S. Henn, A full curvature based algorithm for image registration. J. Math. Imag. Vis. 24(2), 195–208 (2006)
W. Hinterberger, O. Scherzer, Variational methods on the space of functions of bounded Hessian for convexification and denoising. Computing 76(1), 109–133 (2006)
B.K.P. Horn, B.G. Schunck, Determining optical flow. Artif. Intell. 17, 185–203 (1981)
H. Ishikawa, Exact optimization for Markov random fields with convex priors. Pattern. Anal. Mach. Intell. 25(10), 1333–1336 (2003)
E. Laude, T. Möllenhoff, M. Moeller, J. Lellmann, D. Cremers, Sublabel-accurate convex relaxation of vectorial multilabel energies, in European Conference on Computer Vision (Springer, Berlin, 2016), pp. 614–627
J. Lellmann, C. Schnörr, Continuous multiclass labeling approaches and algorithms. SIAM J. Imag. Sci. 4(4), 1049–1096 (2011)
J. Lellmann, J. Kappes, J. Yuan, F. Becker, C. Schnörr, Convex multi-class image labeling by simplex-constrained total variation, in Scale Space and Variational Methods in Computer Vision. Lecture Notes in Computer Science, vol. 5567 (2009), pp. 150–162
J. Lellmann, E. Strekalovskiy, S. Koetter, D. Cremers, Total variation regularization for functions with values in a manifold, in International Conference on Computer Vision (2013), pp. 2944–2951
M. Lysaker, X.C. Tai, Iterative image restoration combining total variation minimization and a second-order functional. Int. J. Comput. Vis. 66(1), 5–18 (2006)
M. Lysaker, A. Lundervold, X.C. Tai, Noise removal using fourth-order partial differential equation with applications to medical magnetic resonance images in space and time. IEEE Trans. Image Process. 12(12), 1579–1590 (2003)
J. Modersitzki, Numerical Methods for Image Registration (Oxford University Press on Demand, Oxford, 2004)
J. Modersitzki, FAIR: Flexible Algorithms for Image Registration (SIAM, Philadelphia, 2009)
T. Möllenhoff, E. Strekalovskiy, M. Moeller, D. Cremers, The primal-dual hybrid gradient method for semiconvex splittings. SIAM J. Imag. Sci. 8(2), 827–857 (2015)
T. Möllenhoff, E. Strekalovskiy, M. Möller, D. Cremers, Low rank priors for color image regularization, in International Workshop on Energy Minimization Methods in Computer Vision and Pattern Recognition (Springer, Berlin, 2015), pp. 126–140
T. Möllenhoff, E. Laude, M. Moeller, J. Lellmann, D. Cremers, Sublabel-accurate relaxation of nonconvex energies, in Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (2016), pp. 3948–3956
M. Nikolova, Analysis of the recovery of edges in images and signals by minimizing nonconvex regularized least-squares. Multiscale Model. Simul. 4(3), 960–991 (2005)
J. Nocedal, S.J. Wright, Numerical Optimization (Springer, Berlin, 2006)
K. Papafitsoros, C.B. Schönlieb, A combined first and second order variational approach for image reconstruction. J. Math. Imag. Vision 48(2), 308–338 (2014)
T. Pock, A. Chambolle, D. Cremers, H. Bischof, A convex relaxation approach for computing minimal partitions, in Computer Vision and Pattern Recognition (2009)
T. Pock, D. Cremers, H. Bischof, A. Chambolle, Global solutions of variational models with convex regularization. J. Imag. Sci. 3(4), 1122–1145 (2010)
R.T. Rockafellar, Convex Analysis (Princeton University Press, Princeton, 1970)
L. Rudin, S. Osher, E. Fatemi, Nonlinear total variation based noise removal algorithms. Physica D 60, 259–268 (1992)
S. Setzer, G. Steidl, T. Teuber, Infimal convolution regularizations with discrete l1-type functionals. Commun. Math. Sci 9(3), 797–872 (2011)
P. Suquet, Existence et régularité des solutions des équations de la plasticité parfaite. C. R. Acad. Sci. Paris, Ser. A 286, 1201–1204 (1978)
J. Yuan, E. Bae, X.C. Tai, Y. Boykov, A continuous max-flow approach to Potts model, in European Conference on Computer Vision (2010), pp. 379–392
C. Zach, D. Gallup, J.M. Frahm, M. Niethammer, Fast global labeling for real-time stereo using multiple plane sweeps, in Vision, Modeling, and Visualization (2008)
C. Zach, C. Häne, M. Pollefeys, What is optimized in convex relaxations for multilabel problems: connecting discrete and continuously inspired map inference. IEEE Trans. Pattern Anal. Mach. Intell. 36(1), 157–170 (2014)
W.P. Ziemer, Weakly Differentiable Functions (Springer, Berlin, 1989)
Acknowledgements
This work was partially funded by the Deutsche Forschungsgemeinschaft (DFG, German Research Foundation)—394737018 “Functional Lifting 2.0: Efficient Convexifications for Imaging and Vision”. We would like to thank Emanuel Laude and Thomas Möllenhoff for providing their library prost, which was used to solve the saddle-point formulation of the problems.
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2018 Springer International Publishing AG, part of Springer Nature
About this paper
Cite this paper
Loewenhauser, B., Lellmann, J. (2018). Functional Lifting for Variational Problems with Higher-Order Regularization. In: Tai, XC., Bae, E., Lysaker, M. (eds) Imaging, Vision and Learning Based on Optimization and PDEs. IVLOPDE 2016. Mathematics and Visualization. Springer, Cham. https://doi.org/10.1007/978-3-319-91274-5_5
Download citation
DOI: https://doi.org/10.1007/978-3-319-91274-5_5
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-319-91273-8
Online ISBN: 978-3-319-91274-5
eBook Packages: Mathematics and StatisticsMathematics and Statistics (R0)