Abstract
Iterative techniques are a well-established tool in modern imaging sciences, allowing to address complex optimization problems via sequences of simpler computational processes. This approach has been significantly expanded in recent years by iterative designs where explicit solutions of optimization subproblems were replaced by black-box applications of ready-to-use modules for denoising or compression. These modular designs are conceptually simple, yet often achieve impressive results. In this chapter, we overview the concept of modular optimization for imaging problems by focusing on structures induced by the alternating direction method of multipliers (ADMM) technique and their applications to intricate restoration and compression problems. We start by emphasizing general guidelines independent of the module type used and only then derive ADMM-based structures relying on denoising and compression methods. The wide perspective on the topic should motivate extensions of the types of problems addressed and the kinds of black boxes utilized by the modular optimization. As an example for a promising research avenue, we present our recent framework employing black-box modules for distributed representations of visual data.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Similar content being viewed by others
References
Afonso, M.V., Bioucas-Dias, J.M., Figueiredo, M.A.: Fast image recovery using variable splitting and constrained optimization. IEEE Trans. Image Process. 19(9), 2345–2356 (2010)
Ahmad, R., Bouman, C.A., Buzzard, G.T., Chan, S., Reehorst, E.T., Schniter, P.: Plug and play methods for magnetic resonance imaging. arXiv preprint arXiv:1903.08616 (2019)
Bahat, Y., Efrat, N., Irani, M.: Non-uniform blind deblurring by reblurring. In: Proceedings of the IEEE International Conference on Computer Vision, pp. 3286–3294 (2017)
Ballé, J., Laparra, V., Simoncelli, E.P.: End-to-end optimized image compression. In: Proceedings of ICLR (2017)
F. Bellard, BPG 0.9.6. [Online]. Available: http://bellard.org/bpg/
Beygi, S., Jalali, S., Maleki, A., Mitra, U.: Compressed sensing of compressible signals. In: IEEE International Symposium on Information Theory (ISIT), pp. 2158–2162 (2017a)
Beygi, S., Jalali, S., Maleki, A., Mitra, U.: An efficient algorithm for compression-based compressed sensing. arXiv preprint arXiv:1704.01992 (2017b)
Boyd, S., Parikh, N., Chu, E., Peleato, B., Eckstein, J.: Distributed optimization and statistical learning via the alternating direction method of multipliers. Found. Trends Mach. Learn. 3(1), 1–122 (2011)
Brifman, A., Romano, Y., Elad, M.: Turning a denoiser into a super-resolver using plug and play priors. In: 2016 IEEE International Conference on Image Processing (ICIP), pp. 1404–1408. IEEE (2016)
Brifman, A., Romano, Y., Elad, M.: Unified single-image and video super-resolution via denoising algorithms. IEEE Trans. Image Process. 28(12), 6063–6076 (2019)
Bruckstein, A.M., Holt, R.J., Netravali, A.N.: Holographic representations of images. IEEE Trans. Image Process. 7(11), 1583–1597 (1998)
Bruckstein, A.M., Holt, R.J., Netravali, A.N.: On holographic transform compression of images. Int. J. Imag. Syst. Technol. 11(5), 292–314 (2000)
Bruckstein, A.M., Ezerman, M.F., Fahreza, A.A., Ling, S.: Holographic sensing. arXiv preprint arXiv:1807.10899 (2018)
Burger, M., Dirks, H., Schonlieb, C.-B.: A variational model for joint motion estimation and image reconstruction. SIAM J. Imag. Sci. 11(1), 94–128 (2018)
Buzzard, G.T., Chan, S.H., Sreehari, S., Bouman, C.A.: Plug-and-play unplugged: optimization-free reconstruction using consensus equilibrium. SIAM J. Imag. Sci. 11(3), 2001–2020 (2018)
Chan, S.H.: Performance analysis of plug-and-play ADMM: a graph signal processing perspective. IEEE Trans. Comput. Imag. 5(2), 274–286 (2019)
Chan, S.H., Wang, X., Elgendy, O.A.: Plug-and-play ADMM for image restoration: fixed-point convergence and applications. IEEE Trans. Comput. Imag. 3(1), 84–98 (2017)
Chatterjee, P., Milanfar, P.: Is denoising dead? IEEE Trans. Image Process. 19(4), 895–911 (2009)
Chou, P.A., Lookabaugh, T., Gray, R.M.: Optimal pruning with applications to tree-structured source coding and modeling. IEEE Trans. Inf. Theory 35(2), 299–315 (1989)
Corona, V., Aviles-Rivero, A.I., Debroux, N., Graves, M., Le Guyader, C., Schönlieb, C.-B., Williams, G.: Multi-tasking to correct: motion-compensated mri via joint reconstruction and registration. In: International Conference on Scale Space and Variational Methods in Computer Vision, pp. 263–274. Springer (2019a)
Corona, V., Benning, M., Ehrhardt, M.J., Gladden, L.F., Mair, R., Reci, A., Sederman, A.J., Reichelt, S., Schönlieb, C.-B.: Enhancing joint reconstruction and segmentation with non-convex bregman iteration. Inverse Probl. 35(5), 055001 (2019b)
Dabov, K., Foi, A., Katkovnik, V., Egiazarian, K.: Image denoising by sparse 3-D transform-domain collaborative filtering. IEEE Trans. Image Process. 16(8), 2080–2095 (2007)
Dar, Y., Bruckstein, A.M.: Benefiting from duplicates of compressed data: shift-based holographic compression of images. J. Math. Imag. Vis. 1–14 63, 380–393 (2021)
Dar, Y., Bruckstein, A.M., Elad, M.: Image restoration via successive compression. In: Picture Coding Symposium (PCS), pp. 1–5 (2016a)
Dar, Y., Bruckstein, A.M., Elad, M., Giryes, R.: Postprocessing of compressed images via sequential denoising. IEEE Trans. Image Process. 25(7), 3044–3058 (2016b)
Dar, Y., Elad, M., Bruckstein, A.M.: Compression for multiple reconstructions. In: IEEE International Conference on Image Processing (ICIP), pp. 440–444 (2018a)
Dar, Y., Elad, M., Bruckstein, A.M.: Optimized pre-compensating compression. IEEE Trans. Image Process. 27(10), 4798–4809 (2018b)
Dar, Y., Elad, M., Bruckstein, A.M.: Restoration by compression. IEEE Trans. Sig. Process. 66(22), 5833–5847 (2018c)
Dar, Y., Elad, M., Bruckstein, A.M.: System-aware compression. In: IEEE International Symposium on Information Theory (ISIT), pp. 2226–2230 (2018d)
Hong, T., Romano, Y., Elad, M.: Acceleration of red via vector extrapolation. J. Vis. Commun. Image Represent. 63, 102575 (2019)
Kamilov, U.S., Mansour, H., Wohlberg, B.: A plug-and-play priors approach for solving nonlinear imaging inverse problems. IEEE Sig. Process. Lett. 24(12), 1872–1876 (2017)
Kwan, C., Choi, J., Chan, S., Zhou, J., Budavari, B.: A super-resolution and fusion approach to enhancing hyperspectral images. Remote Sens. 10(9), 1416 (2018)
Lai, W.-S., Huang, J.-B., Hu, Z., Ahuja, N., Yang, M.-H.: A comparative study for single image blind deblurring. In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, pp. 1701–1709 (2016)
Laparra, V., Berardino, A., Ballé, J., Simoncelli, E.P.: Perceptually optimized image rendering. J. Opt. Soc. Am. A 34, 1511 (2017)
Liu, J., Moulin, P.: Complexity-regularized image denoising. IEEE Trans. Image Process. 10(6), 841–851 (2001)
Moulin, P., Liu, J.: Statistical imaging and complexity regularization. IEEE Trans. Inf. Theory 46(5), 1762–1777 (2000)
Natarajan, B.K.: Filtering random noise from deterministic signals via data compression. IEEE Trans. Sig. Process. 43(11), 2595–2605 (1995)
Ono, S.: Primal-dual plug-and-play image restoration. IEEE Sig. Process. Lett. 24(8), 1108–1112 (2017)
Ortega, A., Ramchandran, K.: Rate-distortion methods for image and video compression. IEEE Sig. Process. Mag. 15(6), 23–50 (1998)
Rissanen, J.: MDL denoising. IEEE Trans. Inf. Theory 46(7), 2537–2543 (2000)
Romano, Y., Elad, M., Milanfar, P.: The little engine that could: regularization by denoising (RED). SIAM J. Imag. Sci. 10(4), 1804–1844 (2017)
Rond, A., Giryes, R., Elad, M.: Poisson inverse problems by the plug-and-play scheme. J. Vis. Commun. Image Represent. 41, 96–108 (2016)
Rott Shaham, T., Michaeli, T.: Deformation aware image compression. In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, pp. 2453–2462 (2018)
Shoham, Y., Gersho, A.: Efficient bit allocation for an arbitrary set of quantizers. IEEE Trans. Acoust. Speech Sig. Process. 36(9), 1445–1453 (1988)
Shukla, R., Dragotti, P.L., Do, M.N., Vetterli, M.: Rate-distortion optimized tree-structured compression algorithms for piecewise polynomial images. IEEE Trans. Image Process. 14(3), 343–359 (2005)
Sreehari, S., Venkatakrishnan, S., Wohlberg, B., Buzzard, G.T., Drummy, L.F., Simmons, J.P., Bouman, C.A.: Plug-and-play priors for bright field electron tomography and sparse interpolation. IEEE Trans. Comput. Imag. 2(4), 408–423 (2016)
Sullivan, G.J., Wiegand, T.: Rate-distortion optimization for video compression. IEEE Sig. Process. Mag. 15(6), 74–90 (1998)
Sullivan, G.J., Ohm, J., Han, W.-J., Wiegand, T.: Overview of the high efficiency video coding (HEVC) standard. IEEE Trans. Circuits Syst. Video Technol. 22(12), 1649–1668 (2012)
Sun, Y., Wohlberg, B., Kamilov, U.S.: An online plug-and-play algorithm for regularized image reconstruction. IEEE Trans. Comput. Imag.5, 395–408 (2019a)
Sun, Y., Xu, S., Li, Y., Tian, L., Wohlberg, B., Kamilov, U.S.: Regularized fourier ptychography using an online plug-and-play algorithm. In: ICASSP 2019-2019 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), pp. 7665–7669. IEEE (2019b)
Tirer, T., Giryes, R.: Image restoration by iterative denoising and backward projections. IEEE Trans. Image Process. 28(3), 1220–1234 (2018a)
Tirer, T., Giryes, R.: An iterative denoising and backwards projections method and its advantages for blind deblurring. In: 2018 25th IEEE International Conference on Image Processing (ICIP), pp. 973–977. IEEE (2018b)
Tirer, T., Giryes, R.: Back-projection based fidelity term for ill-posed linear inverse problems. arXiv preprint arXiv:1906.06794 (2019)
Venkatakrishnan, S.V., Bouman, C.A., Wohlberg, B.: Plug-and-play priors for model based reconstruction. In: IEEE GlobalSIP (2013)
Yazaki, Y., Tanaka, Y., Chan, S.H.: Interpolation and denoising of graph signals using plug-and-play ADMM. In: ICASSP 2019-2019 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), pp. 5431–5435. IEEE (2019)
Zoran, D., Weiss, Y.: From learning models of natural image patches to whole image restoration. In: IEEE International Conference on Computer Vision (ICCV), pp. 479–486 (2011)
Author information
Authors and Affiliations
Corresponding authors
Editor information
Editors and Affiliations
Appendix: Operational Rate-Distortion Optimizations in Block-Based Architectures
Appendix: Operational Rate-Distortion Optimizations in Block-Based Architectures
The computational challenge of operational rate-distortion optimizations (see section “Preliminaries: Lossy Compression via Operational Rate-Distortion Optimization”) is often addressed via the squared-error metric
leading to practical forms of the Lagrangian rate-distortion optimization (25). These useful structures also process the signal x based on its segmentation into a set of nonoverlapping blocks \( \left \lbrace {\mathbf {x}}_i \right \rbrace _{i\in \mathcal {I}}\); here, each of them is a column vector of Nb samples, and \( \mathcal {I} \) is the set of indices corresponding to the nonoverlapping blocks of the signal. Correspondingly, the decompressed signal v is decomposed into a set of nonoverlapping blocks \( \left \lbrace {\mathbf {v}}_i \right \rbrace _{i\in \mathcal {I}}\). This lets us casting (47) into
exhibiting that, for squared-error measures, the total distortion can be computed as the sum of distortions associated with its nonoverlapping blocks. While this property is satisfied for any segmentation of the signal into nonoverlapping blocks, we will exemplify it here for blocks of equal sizes that allow using one block-level compression procedure for all the blocks.
Mirroring the definitions described in section “Preliminaries: Lossy Compression via Operational Rate-Distortion Optimization” for full-signal compression architectures, the block-level process corresponds to a function \( C_b: \mathbb {R}^{N_b} \rightarrow \mathcal {B}_b \), mapping the Nb-dimensional signal-block domain to a discrete set \( \mathcal {B}_b \) of binary compressed representations of blocks. The associated block decompression process is denoted by the function \( F_b: \mathcal {B}_b \rightarrow \mathcal {S}_b \), mapping the binary compressed representations in \( \mathcal {B}_b \) to their decompressed signal blocks from the discrete set \( \mathcal {S}_b \subset \mathbb {R}^{N_b} \). The bit-cost evaluation function \( R_b\left ( {\mathbf {v}}_i \right ) \) is defined to quantify the number of bits needed for the compressed representation matching the decompressed signal block \( {\mathbf {v}}_i \in \mathbb {R}^{N_b} \). Then, the compression of the nonoverlapping signal blocks \( \left \lbrace {\mathbf {x}}_i \right \rbrace _{i\in \mathcal {I}}\) producing the decompressed blocks \( \left \lbrace {\mathbf {v}}_i \right \rbrace _{i\in \mathcal {I}}\) requires a total bit budget satisfying
Plugging the block-based compression design into the Lagrangian form (25) gives
that reduces to a set of block-level rate-distortion Lagrangian optimizations, i.e.,
Note that the block-level optimizations in (51) are independent and refer to the same Lagrangian multiplier λ. Commonly, compression designs are based on processing of low-dimensional blocks, allowing to practically address the block-level optimizations in (51). For example, one can evaluate the Lagrangian cost for all the elements in \( \mathcal {S}_b \) (since this set is sufficiently small).
Rights and permissions
Copyright information
© 2023 Springer Nature Switzerland AG
About this entry
Cite this entry
Dar, Y., Bruckstein, A.M. (2023). Modular ADMM-Based Strategies for Optimized Compression, Restoration, and Distributed Representations of Visual Data. In: Chen, K., Schönlieb, CB., Tai, XC., Younes, L. (eds) Handbook of Mathematical Models and Algorithms in Computer Vision and Imaging. Springer, Cham. https://doi.org/10.1007/978-3-030-98661-2_71
Download citation
DOI: https://doi.org/10.1007/978-3-030-98661-2_71
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-030-98660-5
Online ISBN: 978-3-030-98661-2
eBook Packages: Mathematics and StatisticsReference Module Computer Science and Engineering