• Corpus ID: 247839681

Plug-and-Play Methods for Integrating Physical and Learned Models in Computational Imaging

  title={Plug-and-Play Methods for Integrating Physical and Learned Models in Computational Imaging},
  author={Ulugbek S. Kamilov and Charles A. Bouman and Gregery T. Buzzard and Brendt Wohlberg},
Plug-and-Play Priors (PnP) is one of the most widely-used frameworks for solving computational imaging problems through the integration of physical models and learned models. PnP leverages high-fidelity physical sensor models and powerful machine learning methods for prior modeling of data to provide state-of-the-art reconstruction algorithms. PnP algorithms alternate between minimizing a data-fidelity term to promote data consistency and imposing a learned regularizer in the form of an image… 
Online Deep Equilibrium Learning for Regularization by Denoising
This work theoretically analyze ODER giving insights into its convergence and ability to approximate the traditional DEQ approach, and numerical results suggest the potential improvements in training/testing complexity due to ODER on three distinct imaging applications.
Compressive Ptychography using Deep Image and Generative Priors
A generative model combining deep image priors with deep generative neural network to create a solution for a given dataset is proposed and improved reconstruction accuracy is demonstrated compared to the state-of-the-art method.
Stable and memory-efficient image recovery using monotone operator learning (MOL)
A monotone deep equilibrium learning framework for large-scale inverse problems in imaging based on forwardbackward splitting, where each iteration consists of a gradient descent involving the score function and a conjugate gradient algorithm to encourage data consistency.


Plug-and-Play priors for model based reconstruction
This paper demonstrates with some simple examples how Plug-and-Play priors can be used to mix and match a wide variety of existing denoising models with a tomographic forward model, thus greatly expanding the range of possible problem solutions.
An Online Plug-and-Play Algorithm for Regularized Image Reconstruction
A new online PnP algorithm based on the proximal gradient method (PGM) is introduced, which uses only a subset of measurements at every iteration, which makes it scalable to very large datasets.
A Plug-and-Play Priors Approach for Solving Nonlinear Imaging Inverse Problems
This letter develops the fast iterative shrinkage/thresholding algorithm variant of PPP for model-based nonlinear inverse scattering and shows that the PPP approach is applicable beyond linear inverse problems.
Plug-and-Play Methods for Magnetic Resonance Imaging: Using Denoisers for Image Recovery
This article describes the use of plug-and-play (PnP) algorithms for MRI image recovery and describes how the result of the PnP method can be interpreted as a solution to an equilibrium equation, allowing convergence analysis from this perspective.
Plug-and-Play Priors for Bright Field Electron Tomography and Sparse Interpolation
This paper presents an algorithm for electron tomographic reconstruction and sparse image interpolation that exploits the nonlocal redundancy in images, and demonstrates that the algorithm produces higher quality reconstructions on both simulated and real electron microscope data, along with improved convergence properties compared to other methods.
A Fast Stochastic Plug-and-Play ADMM for Imaging Inverse Problems
This work proposes a new stochastic PnP-ADMM method which is based on introducing Stochastic gradient descent inner-loops within an inexact ADMM framework and provides the theoretical guarantee on the fixed-point convergence for the algorithm under standard assumptions.
Recovery Analysis for Plug-and-Play Priors using the Restricted Eigenvalue Condition
This work shows how to establish theoretical recovery guarantees for PnP/RED by assuming that the solution of these methods lies near the fixedpoints of a deep neural network, and presents numerical results comparing the recovery performance of Pp/RED in compressive sensing against that of recentCompressive sensing algorithms based on generative models.
Scalable Plug-and-Play ADMM With Convergence Guarantees
This work theoretically analyze the convergence of the PnP-ADMM algorithm under a set of explicit assumptions, extending recent theoretical results in the area, and shows the effectiveness of the algorithm with nonsmooth data-fidelity terms and deep neural net priors, its fast convergence compared to existing PNP algorithms, and its scalability in terms of speed and memory.
Joint ptycho-tomography with deep generative priors
This work extends the ADMM framework with plug-and-play (PnP) denoisers by replacing the regularization subproblem with a general denoising operator based on machine learning, and proposes a denoiser parameter to control the effect of the denoisers and to accelerate the solution.
Plug-and-Play Unplugged: Optimization Free Reconstruction using Consensus Equilibrium
Consensus Equilibrium (CE) is introduced, which generalizes regularized inversion to include a much wider variety of both forward components and prior components without the need for either to be expressed with a cost function.