The Little Engine that Could: Regularization by Denoising (RED)
From MaRDI portal
Publication:148569
DOI10.1137/16M1102884zbMATH Open1401.62101arXiv1611.02862MaRDI QIDQ148569FDOQ148569
Authors: Yaniv Romano, Michael Elad, Peyman Milanfar, Yaniv Romano, Peyman Milanfar, Michael Elad 
Publication date: 9 November 2016
Published in: SIAM Journal on Imaging Sciences (Search for Journal in Brave)
Abstract: Removal of noise from an image is an extensively studied problem in image processing. Indeed, the recent advent of sophisticated and highly effective denoising algorithms lead some to believe that existing methods are touching the ceiling in terms of noise removal performance. Can we leverage this impressive achievement to treat other tasks in image processing? Recent work has answered this question positively, in the form of the Plug-and-Play Prior () method, showing that any inverse problem can be handled by sequentially applying image denoising steps. This relies heavily on the ADMM optimization technique in order to obtain this chained denoising interpretation. Is this the only way in which tasks in image processing can exploit the image denoising engine? In this paper we provide an alternative, more powerful and more flexible framework for achieving the same goal. As opposed to the method, we offer Regularization by Denoising (RED): using the denoising engine in defining the regularization of the inverse problem. We propose an explicit image-adaptive Laplacian-based regularization functional, making the overall objective functional clearer and better defined. With a complete flexibility to choose the iterative optimization procedure for minimizing the above functional, RED is capable of incorporating any image denoising algorithm, treat general inverse problems very effectively, and is guaranteed to converge to the globally optimal result. We test this approach and demonstrate state-of-the-art results in the image deblurring and super-resolution problems.
Full work available at URL: https://arxiv.org/abs/1611.02862
Recommendations
- Regularization by denoising via fixed-point projection (RED-PRO)
- Noiseless regularisation by noise
- Regularization under general noise assumptions
- Regularisation by regular noise
- Regularization independent of the noise level: an analysis of quasi-optimality
- Enhancing linear regularization to treat large noise
Image analysis in multivariate analysis (62H35) Image processing (compression, reconstruction, etc.) in information and communication theory (94A08) Computing methodologies for image processing (68U10)
Cites Work
- BM3D Frames and Variational Image Deblurring
- Nonlinear total variation based noise removal algorithms
- Title not available (Why is that?)
- Distributed optimization and statistical learning via the alternating direction method of multipliers
- The Dantzig selector: statistical estimation when \(p\) is much larger than \(n\). (With discussions and rejoinder).
- Global Image Denoising
- Title not available (Why is that?)
- Title not available (Why is that?)
- Image restoration via simultaneous sparse coding: where structured sparsity meets Gaussian scale mixture
- Perturbation of the eigenvectors of the graph Laplacian: application to image denoising
- Non-local regularization of inverse problems
- Coordinate and subspace optimization methods for linear least squares with non-quadratic regularization
- Symmetrizing smoothing filters
- Regularization on graphs with function-adapted diffusion processes
- Secrets of image denoising cuisine
- Boosting of image denoising algorithms
- Sparse and redundant representations. From theory to applications in signal and image processing.
- From Sparse Solutions of Systems of Equations to Sparse Modeling of Signals and Images
- Title not available (Why is that?)
- Progressive Image Denoising Through Hybrid Graph Laplacian Regularization: A Unified Framework
- Progressive Image Denoising
- A General Framework for Regularized, Similarity-Based Image Restoration
- Symmetric Smoothing Filters From Global Consistency Constraints
- Postprocessing of Compressed Images via Sequential Denoising
- Convolutional neural networks analyzed via convolutional sparse coding
- Higher-Order Feature-Preserving Geometric Regularization
- Generalizing the Nonlocal-Means to Super-Resolution Reconstruction
- Fast Gradient-Based Algorithms for Constrained Total Variation Image Denoising and Deblurring Problems
- Principal Neighborhood Dictionaries for Nonlocal Means Image Denoising
- Is Denoising Dead?
- Fast Image Recovery Using Variable Splitting and Constrained Optimization
- An Augmented Lagrangian Approach to the Constrained Optimization Formulation of Imaging Inverse Problems
- Image Deblurring and Super-Resolution by Adaptive Sparse Domain Selection and Adaptive Regularization
- Patch-Based Near-Optimal Image Denoising
- Solving Inverse Problems With Piecewise Linear Estimators: From Gaussian Mixture Models to Structured Sparsity
- Sparse Representation Based Image Interpolation With Nonlocal Autoregressive Modeling
- How to SAIF-ly Boost Denoising Performance
- Nonlocally Centralized Sparse Representation for Image Restoration
- Title not available (Why is that?)
- A Review of Image Denoising Algorithms, with a New One
- Local and nonlocal discrete regularization on weighted graphs for image and mesh processing
- Nonlocal discrete \(p\)-Laplacian driven image and manifold processing
Cited In (66)
- Optimal Dirichlet control of partial differential equations on networks
- Learning Regularization Parameter-Maps for Variational Image Reconstruction Using Deep Neural Networks and Algorithm Unrolling
- Constrained and unconstrained deep image prior optimization models with automatic regularization
- Plug-and-play gradient-based denoisers applied to CT image enhancement
- Explainable bilevel optimization: an application to the Helsinki Deblur Challenge
- An image reconstruction model regularized by edge-preserving diffusion and smoothing for limited-angle computed tomography
- Whiteness constraints in a unified variational framework for image restoration
- Regularising inverse problems with generative machine learning models
- Blind Poissonian image deblurring regularized by a denoiser constraint and deep image prior
- Regularization by denoising via fixed-point projection (RED-PRO)
- On maximum a posteriori estimation with Plug \& Play priors and stochastic gradient descent
- Joint Gaussian dictionary learning and tomographic reconstruction
- Speckle reduction in matrix-log domain for synthetic aperture radar imaging
- Regularization methods in the analysis of a series of scintillation fluorescence microscopy images
- Deep learning for inverse problems. Abstracts from the workshop held March 7--13, 2021 (hybrid meeting)
- Inversion of Band-Limited Discrete Fourier Transforms of Binary Images: Uniqueness and Algorithms
- A support-denoiser-driven framework for single image restoration
- Improving image deblurring
- An Unrolled Implicit Regularization Network for Joint Image and Sensitivity Estimation in Parallel MR Imaging with Convergence Guarantee
- Self-Supervised Deep Learning for Image Reconstruction: A Langevin Monte Carlo Approach
- Computational Imaging Through Atmospheric Turbulence
- Rendition: reclaiming what a black box takes away
- Patch ordering as a regularization for inverse problems in image processing
- A literature survey of matrix methods for data science
- Effective new methods for automated parameter selection in regularized inverse problems
- An introduction to the mathematics of deep learning
- Image Denoising: The Deep Learning Revolution and Beyond—A Survey Paper
- Convolutional proximal neural networks and plug-and-play algorithms
- Regularization theory of the analytic deep prior approach
- Patch-based image restoration using expectation propagation
- Deep synthesis network for regularizing inverse problems
- Denoising AMP for MRI reconstruction: BM3D-AMP-MRI
- RTR-DPD: reweighted Tikohonov regularization for blind deblurring via dual principles of discriminativeness
- Modern regularization methods for inverse problems
- Title not available (Why is that?)
- Solving Inverse Problems by Joint Posterior Maximization with Autoencoding Prior
- Bayesian Imaging with Data-Driven Priors Encoded by Neural Networks
- Compressive Computed Tomography Reconstruction through Denoising Approximate Message Passing
- redR
- Remove the salt and pepper noise based on the high order total variation and the nuclear norm regularization
- NETT: solving inverse problems with deep neural networks
- Designing rotationally invariant neural networks from PDEs and variational methods
- Deep-plug-and-play proximal Gauss-Newton method with applications to nonlinear, ill-posed inverse problems
- Rate-optimal denoising with deep neural networks
- Multi-frame super resolution via deep plug-and-play CNN regularization
- Solving inverse problems using data-driven models
- Accelerating Proximal Markov Chain Monte Carlo by Using an Explicit Stabilized Method
- WPPNets and WPPFlows: The Power of Wasserstein Patch Priors for Superresolution
- Plug-and-Play Unplugged: Optimization-Free Reconstruction Using Consensus Equilibrium
- Joint demosaicking and denoising benefits from a two-stage training strategy
- NF-ULA: normalizing flow-based unadjusted Langevin algorithm for imaging inverse problems
- Convergent data-driven regularizations for CT reconstruction
- Three-operator splitting for learning to predict equilibria in convex games
- Adaptive total variational regularization of Gaussian denoisers for multiplicative noise removal
- Averaged deep denoisers for image regularization
- Preconditioned Plug-and-Play ADMM with Locally Adjustable Denoiser for Image Restoration
- PnP-ReG: Learned Regularizing Gradient for Plug-and-Play Gradient Descent
- Modular ADMM-based strategies for optimized compression, restoration, and distributed representations of visual data
- Learned regularizers for inverse problems
- Iterative gradient denoising algorithm for signal recovery using analysis-based implicit prior
- The mathematics of artificial intelligence
- Normalizing flow regularization for photoacoustic tomography
- Review of diffusion models: theory and applications
- Neural-network-based regularization methods for inverse problems in imaging
- Robustness and exploration of variational and machine learning approaches to inverse problems: an overview
- Learning sparsity-promoting regularizers using bilevel optimization
Uses Software
This page was built for publication: The Little Engine that Could: Regularization by Denoising (RED)
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q148569)
