The Little Engine That Could: Regularization by Denoising (RED)

@article{Romano2017TheLE,
  title={The Little Engine That Could: Regularization by Denoising (RED)},
  author={Yaniv Romano and Michael Elad and Peyman Milanfar},
  journal={ArXiv},
  year={2017},
  volume={abs/1611.02862}
}
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 has led 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 ($P^3$) method, showing that any inverse… 
An Interpretation Of Regularization By Denoising And Its Application With The Back-Projected Fidelity Term
TLDR
This paper shows that the RED gradient can be seen as a (sub)gradient of a prior function—but taken at a denoised version of the point, and proposes to combine RED with the Back-Projection (BP) fidelity term rather than the common Least Squares (LS) term that is used in previous works.
Image Denoising—Old and New
TLDR
This chapter reviews some of the most successful variational approaches for image reconstruction and discusses their structural advantages and disadvantages in comparison to learning-based approaches, and presents a framework to incorporate deep learning approaches in inverse problem formulations.
Regularization by Denoising via Fixed-Point Projection (RED-PRO)
TLDR
This paper reformulates RED as a convex optimization problem utilizing a projection (RED- PRO) onto the fixed-point set of demicontractive denoisers, and establishes a novel unification of RED-PRO and PnP, while providing guarantees for their convergence to the globally optimal solution.
RED-UCATION: A Novel CNN Architecture Based on Denoising Nonlinearities
TLDR
This work describes how a learned version of RED defines a novel convolutional neural network architecture, where the commonly used point-wise nonlinearities are replaced by a denoising engine, and shows how this network can be optimized end- to-end using a back - propagation that relies on guided denoised algorithms.
Regularization by Denoising: Clarifications and New Interpretations
TLDR
This work proposes a new framework called Score-Matching by Denoising (SMD), which aims to match a “score” (i.e., the gradient of a log-prior) and shows tight connections between SMD, kernel density estimation, and constrained minimum mean-squared error denoising.
Stochastic Image Denoising by Sampling from the Posterior Distribution
TLDR
This work proposes a novel stochastic denoising approach that produces viable and high perceptual quality results, while maintaining a small MSE, and presents an extension of the algorithm for handling the inpainting problem, recovering missing pixels while removing noise from partially given data.
Iterative Joint Image Demosaicking and Denoising Using a Residual Denoising Network
TLDR
A novel algorithm that is inspired by powerful classical image regularization methods, large-scale optimization, and deep learning techniques is proposed, which involves a trainable denoising network and has a transparent and clear interpretation compared with other black-box data driven approaches.
Image Restoration by Iterative Denoising and Backward Projections
  • Tom Tirer, R. Giryes
  • Computer Science, Mathematics
    IEEE Transactions on Image Processing
  • 2019
TLDR
An alternative method for solving inverse problems using off-the-shelf denoisers, which requires less parameter tuning and is competitiveness with task-specific techniques and the P&P approach for image inpainting and deblurring is proposed.
Boosting the Performance of Plug-and-Play Priors via Denoiser Scaling
TLDR
A scaling parameter is introduced that adjusts the magnitude of the denoiser input and output so that the performance of plug-and-play prirs for denoising CNN priors that do not have explicitly tunable parameters is improved.
...
...

References

SHOWING 1-10 OF 106 REFERENCES
Is Denoising Dead?
TLDR
This work estimates a lower bound on the mean squared error of the denoised result and compares the performance of current state-of-the-art denoising methods with this bound, showing that despite the phenomenal recent progress in the quality of denoizing algorithms, some room for improvement still remains for a wide class of general images, and at certain signal-to-noise levels.
Boosting of Image Denoising Algorithms
TLDR
This paper introduces an interesting interpretation of the SOS algorithm as a technique for closing the gap between the local patch-modeling and the global restoration task, thereby leading to improved performance.
Natural image denoising: Optimality and inherent bounds
TLDR
This paper takes a non parametric approach and represents the distribution of natural images using a huge set of 1010 patches and derives a simple statistical measure which provides a lower bound on the optimal Bayesian minimum mean square error (MMSE).
How to SAIF-ly Boost Denoising Performance
TLDR
Experiments illustrate that the proposed spatially adaptive iterative filtering (SAIF) strategy can significantly relax the base algorithm's sensitivity to its tuning (smoothing) parameters, and effectively boost the performance of several existing denoising filters to generate state-of-the-art results under both simulated and practical conditions.
Image denoising: Can plain neural networks compete with BM3D?
TLDR
This work attempts to learn this mapping directly with a plain multi layer perceptron (MLP) applied to image patches and shows that by training on large image databases it is able to compete with the current state-of-the-art image denoising methods.
Global Image Denoising
TLDR
Experiments illustrate that the proposed global filter can effectively globalize any existing denoising filters to estimate each pixel using all pixels in the image, hence improving upon the best patch-based methods.
Sparsity-based image denoising via dictionary learning and structural clustering
TLDR
A double-header l1-optimization problem where the regularization involves both dictionary learning and structural structuring is formulated and a new denoising algorithm built upon clustering-based sparse representation (CSR) is proposed.
Improving K-SVD denoising by post-processing its method-noise
TLDR
This paper proposes a novel technique for improving the K-SVD denoising results by recovering the “stolen” image content information from the method-noise by performing iterations of de-noising using the same atoms that represent the first-stage de- noised patch.
Patch-Based Near-Optimal Image Denoising
TLDR
This paper proposes a patch-based Wiener filter that exploits patch redundancy for image denoising that is on par or exceeding the current state of the art, both visually and quantitatively.
Image Denoising Via Sparse and Redundant Representations Over Learned Dictionaries
TLDR
This work addresses the image denoising problem, where zero-mean white and homogeneous Gaussian additive noise is to be removed from a given image, and uses the K-SVD algorithm to obtain a dictionary that describes the image content effectively.
...
...