Plug-and-Play priors for model based reconstruction
The DOI matches its Crossref record; 32 of the 32 references it cites were found in a registry or an index.
Abstract
Model-based reconstruction is a powerful framework for solving a variety of inverse problems in imaging. In recent years, enormous progress has been made in the problem of denoising, a special case of an inverse problem where the forward model is an identity operator. Similarly, great progress has been made in improving model-based inversion when the forward model corresponds to complex physical measurements in applications such as X-ray CT, electron-microscopy, MRI, and ultrasound, to name just a few. However, combining state-of-the-art denoising algorithms (i.e., prior models) with state-of-the-art inversion methods (i.e., forward models) has been a challenge for many reasons. In this paper, we propose a flexible framework that allows state-of-the-art forward models of imaging systems to be matched with state-of-the-art priors or denoising models. This framework, which we term as Plug-and-Play priors, has the advantage that it dramatically simplifies software integration, and moreover, it allows state-of-the-art denoising methods that have no known formulation as an optimization problem to be used. We demonstrate 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.
Overview
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
Citations
Cites
A Threshold Selection Method from Gray-Level Histograms
Otsu, N. (1979) · IEEE Transactions on Systems, Man, and Cybernetics · doi:10.1109/tsmc.1979.4310076 · Cited by 43,398 · cited 1×
“In order to initialize the class labels b we use Otsu’s method [30] on a low pass filtered version of the noisy input y.”
Boyd, S., Parikh, N., Chu, E., Peleato, B., & Eckstein, J. (2011) · Foundations and Trends® in Machine Learning · doi:10.1561/2200000016 · Cited by 17,075
Nonlinear total variation based noise removal algorithms
Rudin, L. I., Osher, S., & Fatemi, E. (1992) · Physica D: Nonlinear Phenomena · doi:10.1016/0167-2789(92)90242-f · Cited by 15,704 · cited 1×
“We compare reconstructions using the Plug-andPlay priors framework by experimenting with six different denoising techniques/priors - K-SVD [4], BM3D [5], PLOW [7], Total Variation (TV) [21], q-GGMRF [22] and discrete reconstruction (DR) [23].”
Boyd, S. (2010) · now publishers, Inc. eBooks · doi:10.1561/9781601984616 · Cited by 12,677
Image denoising by sparse 3-D transform-domain collaborative filtering
Dabov, K., Foi, A., Katkovnik, V., & Egiazarian, K. (2007) · IEEE Transactions on Image Processing · doi:10.1109/tip.2007.901238 · Cited by 9,283 · cited 3×
“We compare reconstructions using the Plug-andPlay priors framework by experimenting with six different denoising techniques/priors - K-SVD [4], BM3D [5], PLOW [7], Total Variation (TV) [21], q-GGMRF [22] and discrete reconstruction (DR) [23].”
A Non-Local Algorithm for Image Denoising
Buades, A., Coll, B., & Morel, J.-M. (2005) · 2005 IEEE Computer Society Conference on Computer Vision and Pattern Recognition (CVPR'05) · doi:10.1109/cvpr.2005.38 · Cited by 7,616 · cited 1×
“Examples of new methods include a wide variety of patch based non-local means approaches [3], generalizations of bi-lateral filtering approaches [1], patch-based dictionary learning methods such as K-SVD [4], block-matching with transform-based denoising such as BM3D [5], and a variety of total-variation [2], and Mark…”
Image Denoising Via Sparse and Redundant Representations Over Learned Dictionaries
Elad, M., & Aharon, M. (2006) · IEEE Transactions on Image Processing · doi:10.1109/tip.2006.881969 · Cited by 5,321 · cited 3×
“We compare reconstructions using the Plug-and- Play priors framework by experimenting with six different denoising techniques/priors -K-SVD [4], BM3D [5], PLOW [7], Total Variation (TV) [21], q-GGMRF [22] and discrete reconstruction (DR) [23].”
The Split Bregman Method for L1-Regularized Problems
Goldstein, T., & Osher, S. (2009) · SIAM Journal on Imaging Sciences · doi:10.1137/080725891 · Cited by 4,282 · cited 3×
“Examples of new methods include a wide variety of patch based non-local means approaches [3], generalizations of bi-lateral filtering approaches [1], patch-based dictionary learning methods such as K-SVD [4], block-matching with transform-based denoising such as BM3D [5], and a variety of total-variation [2], and Mark…”
Fast Image Recovery Using Variable Splitting and Constrained Optimization
Afonso, M. V., Bioucas-Dias, J. M., & Figueiredo, M. A. T. (2010) · IEEE Transactions on Image Processing · doi:10.1109/tip.2010.2047910 · Cited by 1,223 · cited 3×
“In order to design an algorithm for (3) that decouples the forward and prior terms, we first split the variable x into two new variables x and v, and reformulate equation (3) as the following constrained optimization problem [15].”
InfluentialMR Image Reconstruction From Highly Undersampled k-Space Data by Dictionary Learning
Ravishankar, S., & Bresler, Y. (2011) · IEEE Transactions on Medical Imaging · doi:10.1109/tmi.2010.2090538 · Cited by 1,059 · cited 2×
“Model-based reconstruction is a powerful framework for solving a variety of inverse problems in imaging including denoising, deblurring, tomographic reconstruction, and MRI reconstruction.”
Cited by
Items 201–300 of 1,000
Hyperspectral Image Denoising via Low-Rank Representation and CNN Denoiser
Sun, H., Liu, M., Zheng, K., Yang, D., Li, J., & Gao, L. (2021) · IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing · doi:10.1109/jstars.2021.3138564 · Cited by 35
The Practicality of Stochastic Optimization in Imaging Inverse Problems
Tang, J., Egiazarian, K., Golbabaee, M., & Davies, M. E. (2020) · IEEE Transactions on Computational Imaging · doi:10.1109/tci.2020.3032101 · Cited by 41
“c or stochastic proximal gradient methods. Our results can also be extended for understanding and analyzing the limitations of stochastic gradient-based methods [57, 56] with the plug-and-play priors [64, 30] and regularization-by-denoising schemes [53, 51] in imaging inverse problems, which we leave as a future direc…”
Asymptotic bias of inexact Markov chain Monte Carlo methods in high dimension
Durmus, A., & Eberle, A. (2024) · The Annals of Applied Probability · doi:10.1214/23-aap2034 · Cited by 51
A 3-D Sparse SAR Imaging Method Based on Plug-and-Play
Wang, Y., He, Z., Zhan, X., Zeng, Q., & Hu, Y. (2022) · IEEE Transactions on Geoscience and Remote Sensing · doi:10.1109/tgrs.2022.3221934 · Cited by 37
Zhang, R., Ye, D. H., Pal, D., Thibault, J., Sauer, K. D., & Bouman, C. A. (2016) · IEEE Transactions on Computational Imaging · doi:10.1109/tci.2016.2582042 · Cited by 35
“Though effective in denoising applications, other patch-based methods such as BM3D [29] are not directly suited for application in model-based reconstruction problems, and integration of such methods into iterative reconstruction is still an active topic of research [30]– [32].”
Self-citationPlug-and-Play gradient-based denoisers applied to CT image enhancement
Cascarano, P., Piccolomini, E. L., Morotti, E., & Sebastiani, A. (2022) · Applied Mathematics and Computation · doi:10.1016/j.amc.2022.126967 · Cited by 39
Liu, X., Chen, Y., Peng, Z., Wu, J., & Wang, Z. (2018) · Applied Sciences · doi:10.3390/app8101864 · Cited by 32
Robust Compressive Phase Retrieval via Deep Generative PriorsPreprint
Shamshad, F., & Ahmed, A. M. (2018) · arXiv (Cornell University) · doi:10.48550/arxiv.1808.05854 · Cited by 32
A Novel Sharpening Approach for Superresolving Multiresolution Optical Images
Paris, C., Bioucas‐Dias, J. M., & Bruzzone, L. (2018) · IEEE Transactions on Geoscience and Remote Sensing · doi:10.1109/tgrs.2018.2867284 · Cited by 31
“whose solution is the so-called proximity operator [45] of φλ/μ, denoted as proxφλ/μ, computed at (z− d2).”
Hyperspectral image restoration via CNN denoiser prior regularized low-rank tensor recovery
Zeng, H., Xie, X., Cui, H., Zhao, Y., & Ning, J. (2020) · Computer Vision and Image Understanding · doi:10.1016/j.cviu.2020.103004 · Cited by 41
Multi-Slice Fusion for Sparse-View and Limited-Angle 4D CT Reconstruction
Majee, S., Balke, T., Kemp, C. A. J., Buzzard, G. T., & Bouman, C. A. (2021) · IEEE Transactions on Computational Imaging · doi:10.1109/tci.2021.3074881 · Cited by 31
“Consequently, PnP algorithms have been created as a method for representing prior models as denoising operations [13], [14].”
Self-citationInverse Problem of Ultrasound Beamforming With Denoising-Based Regularized Solutions
Goudarzi, S., Basarab, A., & Rivaz, H. J. (2022) · IEEE Transactions on Ultrasonics Ferroelectrics and Frequency Control · doi:10.1109/tuffc.2022.3198874 · Cited by 31
Wasserstein Distance-Based Full-Waveform Inversion With a Regularizer Powered by Learned Gradient
Yang, F., & Ma, J. (2023) · IEEE Transactions on Geoscience and Remote Sensing · doi:10.1109/tgrs.2023.3241723 · Cited by 31
“Plug-and-play priors (PnP) [27] and regularization by”
Bian, W., Jang, A., & Liu, F. (2024) · Magnetic Resonance in Medicine · doi:10.1002/mrm.30045 · Cited by 42
“…operator for regularization 𝜆 ( z ) is defined as: prox 𝜆 ,𝜌 ( y ) ∶= arg min This step involves implementing the proximal operation for regularization , which is equivalent to finding the maximum-a-posteriori solution for the Gaussian denoising problem at a noise level √ 𝜆 ∕ 𝜌 .”
Provably Convergent Plug-and-Play Quasi-Newton Methods
Tan, H. Y., Mukherjee, S., Tang, J., & Schönlieb, C. (2024) · SIAM Journal on Imaging Sciences · doi:10.1137/23m157185x · Cited by 42
Hyperspectral image denoising and anomaly detection based on low-rank and sparse representations
Zhuang, L., Gao, L., Zhang, B., & Bioucas‐Dias, J. M. (2017) · Image and signal processing for remote sensing. Conference · doi:10.1117/12.2280456 · Cited by 30
MemXCT: memory-centric X-ray CT reconstruction with massive parallelization
Hidayetoğlu, M., Biçer, T., de Gonzalo, S. G., Ren, B., Gürsoy, D., Kettimuthu, R., Foster, I. T., & Hwu, W. (2019) · International Conference for High Performance Computing, Networking, Storage and Analysis (SC) · doi:10.1145/3295500.3356220 · Cited by 40
Distributed Optimization for Nonrigid Nano-Tomography
Nikitin, V., Andrade, V. D., Slyamov, A., Gould, B., Zhang, Y., Sampathkumar, V., Kasthuri, N., Gursoy, D., & Carlo, F. D. (2021) · IEEE Transactions on Computational Imaging · doi:10.1109/tci.2021.3060915 · Cited by 30
“possible to prepare a general Plug-and-Play ADMM framework (similar to the one proposed in [86]) for denoising, where state-of-the-art learning based methods can be applied into the”
NTIRE 2021 Challenge on Burst Super-Resolution: Methods and Results
Bhat, G., Danelljan, M., Timofte, R., Akita, K., Cho, W., Fan, H., Jia, L., Kim, D., Lecouat, B., Li, Y., et al. (2021) · IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW) · doi:10.1109/cvprw53098.2021.00073 · Cited by 44
Petascale XCT: 3D Image Reconstruction with Hierarchical Communications on Multi-GPU Nodes
Hidayetoğlu, M., Biçer, T., de Gonzalo, S. G., Ren, B., Andrade, V. D., Gürsoy, D., Kettimuthu, R., Foster, I. T., & Hwu, W. (2020) · International Conference for High Performance Computing, Networking, Storage and Analysis (SC) · doi:10.1109/sc41405.2020.00041 · Cited by 39
Learned Regularizations for Multi‐Parameter Elastic Full Waveform Inversion Using Diffusion Models
Taufik, M. H., Wang, F., & Alkhalifah, T. A. (2024) · Journal of Geophysical Research Machine Learning and Computation · doi:10.1029/2024jh000125 · Cited by 29
Huang, J., Wu, Y., Wang, F., Fang, Y., Yang, N., Alkan, C., Abraham, D. P., Liao, C., Xu, L., Gao, Z., et al. (2024) · IEEE Reviews in Biomedical Engineering · doi:10.1109/rbme.2024.3485022 · Cited by 37
“Plug-and-Play (PnP) approaches [61] ( Figure 4 ) in MRI reconstruction utilize advanced algorithms as modular components substituted (i.e., “plugged”) into an iterative process, without extensive modification of the reconstruction pipeline [23].”
Graph Signal Restoration Using Nested Deep Algorithm Unrolling
Nagahama, M., Yamada, K., Tanaka, Y., Chan, S. H., & Eldar, Y. C. (2022) · IEEE Transactions on Signal Processing · doi:10.1109/tsp.2022.3180546 · Cited by 43
“Second, we propose a nested DAU based on unrolling the iterations of Plug-and-Play ADMM (PnP-ADMM) [4], [5], [42][44].”
Preconditioned Plug-and-Play ADMM with Locally Adjustable Denoiser for Image Restoration
Pendu, M. L., & Guillemot, C. M. (2023) · SIAM Journal on Imaging Sciences · doi:10.1137/22m1504809 · Cited by 36
Hyperspectral Image Denoising via Robust Subspace Estimation and Group Sparsity Constraint
Fu, X., Guo, Y., Xu, M., & Jia, S. (2023) · IEEE Transactions on Geoscience and Remote Sensing · doi:10.1109/tgrs.2023.3277832 · Cited by 34
“hand-crafted regularization φ(V2) [40], [41], [58].”
Integrating self-supervised denoising in inversion-based seismic deblending
Luiken, N., Ravasi, M., & Birnie, C. E. (2023) · Geophysics · doi:10.1190/geo2023-0131.1 · Cited by 29
Full-Waveform Inversion Using a Learned Regularization
Sun, P., Yang, F., Liang, H., & Ma, J. (2023) · IEEE Transactions on Geoscience and Remote Sensing · doi:10.1109/tgrs.2023.3322964 · Cited by 42
“Recently, a new type of method, called plug-and-play priors (PnP) [8], has presented a powerful ability to solve regularized inverse problems using more sophisticated denoisers as priors within iterative inversion algorithms.”
Wu, E. G., Brackbill, N., Rhoades, C. E., Kling, A., Gogliettino, A. R., Shah, N. P., Sher, A., Litke, A. M., Simoncelli, E. P., & Chichilnisky, E. J. (2024) · Nature Communications · doi:10.1038/s41467-024-52304-7 · Cited by 29
“…likelihood optimization step (solved with unconstrained convex minimization) and a prior optimization step (solved with a single forward pass of the denoiser [6, 20]) (Figures 1d,e, see Methods), yielding an estimate of the most probable image given the RGC spikes and natural image statistics.”
Online Regularization by Denoising with Applications to Phase Retrieval
Wu, Z., Sun, Y., Liu, J., & Kamilov, U. S. (2019) · IEEE International Conference on Computer Vision Workshops · doi:10.1109/iccvw.2019.00482 · Cited by 41
Boosting the Performance of Plug-and-Play Priors via Denoiser Scaling
Xu, X., Liu, J., Sun, Y., Wohlberg, B. E., & Kamilov, U. S. (2020) · Asilomar Conference on Signals, Systems and Computers (ACSSC) · doi:10.1109/ieeeconf51394.2020.9443410 · Cited by 30
“The observation that the proximal operator is an image denoiser for AWGN prompted the development of PnP [15], where the operator proxτh(·), within a proximal algorithm, is replaced with a more general image denoiser D(·), such as BM3D [31] or DnCNN [32].”
Self-citationFablet, R., Drumetz, L., & Rousseau, F. (2020) · arXiv (Cornell University) · doi:10.48550/arxiv.2006.03653 · Cited by 38
Rotation Equivariant Proximal Operator for Deep Unfolding Methods in Image Restoration
Fu, J., Xie, Q., Meng, D., & Xu, Z. (2024) · IEEE Transactions on Pattern Analysis and Machine Intelligence · doi:10.1109/tpami.2024.3383532 · Cited by 38
“TheearlyPnPalgorithmscanbetracedbackto[21],[22],[23], [24], [25].”
Investigating Task-Driven Latent Feasibility for Nonconvex Image Modeling
Liu, R., Mu, P., Chen, J., Fan, X., & Luo, Z. (2020) · IEEE Transactions on Image Processing · doi:10.1109/tip.2020.3004733 · Cited by 31
“Similar to this view, plug-in schemes replace the regularization term by using taskrelated operator and have been extensively studied with a great empirical success for vision problems [24], [8], [25], [26], [27].”
Deep prior-based sparse representation model for diffraction imaging: A plug-and-play method
Shi, B., Lian, Q., & Chang, H. (2019) · Signal Processing · doi:10.1016/j.sigpro.2019.107350 · Cited by 31
Deep Spatial-Angular Regularization for Compressive Light Field Reconstruction over Coded Apertures
Guo, M., Hou, J., Jin, J., Chen, J., & Chau, L. (2020) · Lecture notes in computer science · doi:10.1007/978-3-030-58536-5_17 · Cited by 28
Feasibility-based fixed point networks
Heaton, H., Fung, S. W., Gibali, A., & Yin, W. (2021) · Fixed Point Theory and Algorithms for Sciences and Engineering · doi:10.1186/s13663-021-00706-3 · Cited by 44
Xiao, C., Liu, T., Ying, X., Wang, Y., Li, M., Liu, L., An, W., & Chen, Z. (2023) · IEEE Transactions on Geoscience and Remote Sensing · doi:10.1109/tgrs.2023.3243055 · Cited by 31
“Unlike traditional model-based methods that require explicit and handcrafted image priors, model-based methods with deep priors [33], [34], [35] can incorporate implicit deep priors from deep CNN networks for image restoration [36].”
Self-supervised Scalable Deep Compressed Sensing
Chen, B., Zhang, X., Liu, S., Zhang, Y., & Zhang, J. (2024) · International Journal of Computer Vision · doi:10.1007/s11263-024-02209-1 · Cited by 35
“Some plug-and-play (PnP) (Venkatakrishnan et al., 2013; Sun et al., 2019; Ryu et al., 2019; Kadkhodaie & Simoncelli, 2021; Kamilov et al., 2023), regularization by denoising (RED) (Romano et al., 2017; Liu et al., 2022a), and diffusion model-based (Kawar et al., 2022; Wang et al., 2023b; Chung et…”
Zeng, H., Xie, X., Kong, W., Cui, S., & Ning, J. (2020) · IEEE Access · doi:10.1109/access.2020.2979809 · Cited by 28
“Fortunately, a recent Plug-and-Play (PnP) framework [28] has reported empirical success on a large variety of image processing [1], [2], [29], [30].”
Unfolded Proximal Neural Networks for Robust Image Gaussian Denoising
Le, H. T. V., Repetti, A., & Pustelnik, N. (2024) · IEEE Transactions on Image Processing · doi:10.1109/tip.2024.3437219 · Cited by 35
“During the last decade, the performances of proximal algorithms have been pushed to the next level by mixing them with deep learning approaches [19], [20], [21] leading to PNNs that consists in unrolling optimisation algorithms over a fixed number of iterations [22], [23].”
Iterative Residual CNNs for Burst Photography Applications
Kokkinos, F., & Lefkimmiatis, S. (2019) · IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings - IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings · doi:10.1109/cvpr.2019.00608 · Cited by 31
“Another important point we would like to highlight is that our approach, as opposed to other related methods that use a network to replace the proximal operator such as IRCNN [43], Plug and Play [40] and RED [35], is completely parameter-free and, thus, no manual tuning is required so that a good reconstruction is pro…”
Photon Limited Non-Blind Deblurring Using Algorithm Unrolling
Sanghvi, Y., Gnanasambandam, A., & Chan, S. H. (2022) · IEEE Transactions on Computational Imaging · doi:10.1109/tci.2022.3209939 · Cited by 34
“In the Plug-and-Play framework [16], [17], the z update in (8b) is implemented by an image denoiser.”
A Bayesian Based Deep Unrolling Algorithm for Single-Photon Lidar Systems
Koo, J., Halimi, A., & McLaughlin, S. (2022) · IEEE Journal of Selected Topics in Signal Processing · doi:10.1109/jstsp.2022.3170228 · Cited by 27
“The parameters of the resulting models are then estimated using optimization [9]–[14], Markov chain Monte-Carlo (MCMC) [8], [15], [16], marginalization [17], expectation-maximization [18], [19], or Plug-and-Play methods [20], [21].”
On Maximum a Posteriori Estimation with Plug & Play Priors and Stochastic Gradient Descent
Laumont, R., Bortoli, V. D., Almansa, A., Delon, J., Durmus, A., & Pereyra, M. (2023) · Journal of Mathematical Imaging and Vision · doi:10.1007/s10851-022-01134-7 · Cited by 38
“Following the seminal work [60], this kind of splitting schemes have become ubiquitous in cases where U (and hence D† ε) are unknown and unspecified, but a denoiser Dε is available and assumed to be a good approximation of D † ε = proxεU .”
Deep Equilibrium Learning of Explicit Regularization Functionals for Imaging Inverse Problems
Zou, Z., Liu, J., Wohlberg, B. E., & Kamilov, U. S. (2023) · IEEE Open Journal of Signal Processing · doi:10.1109/ojsp.2023.3296036 · Cited by 40
Self-citationHu, Y., Gan, W., Ying, C., Wang, T., Eldeniz, C., Liu, J., Chen, Y., An, H., & Kamilov, U. S. (2024) · Magnetic Resonance in Medicine · doi:10.1002/mrm.30121 · Cited by 34
Solving Inverse Computational Imaging Problems Using Deep Pixel-Level Prior
Dave, A., Vadathya, A. K., Subramanyam, R., Baburajan, R., & Mitra, K. (2018) · IEEE Transactions on Computational Imaging · doi:10.1109/tci.2018.2882698 · Cited by 28
Integrating Local and Non-local Denoiser Priors for Image Restoration
Gu, S., Timofte, R., & Gool, L. V. (2018) · Proceedings - International Conference on Pattern Recognition/Proceedings/International Conference on Pattern Recognition · doi:10.1109/icpr.2018.8545043 · Cited by 30
“Early works [22], [23], [25] tend to utilize the classical BM3D denoiser [10], which take benefit from the NSS prior for denoising.”
InverseBench: Benchmarking Plug-and-Play Diffusion Priors for Inverse Problems in Physical Sciences
Zheng, H., Chu, W., Zhang, B., Wu, Z., Wang, A., Feng, B. T., Zou, C., Sun, Y., Kovachki, N., Ross, Z. E., et al. (2025) · International Conference on Learning Representations · doi:10.48550/arxiv.2503.11043 · Cited by 84
“We use the term Plug-and-Play Diffusion Prior (PnPDP) to refer to the class of recent methods that use diffusion models (or the denoising network within) as plug-and-play priors (Venkatakrishnan et al., 2013) for solving inverse problems.”
Hyperspectral Image Mixed Noise Removal Using Subspace Representation and Deep CNN Image Prior
Zhuang, L., Ng, M. K., & Fu, X. (2021) · Remote Sensing · doi:10.3390/rs13204098 · Cited by 30
“These fall into the area of research called the plug-and-play technique [29,30] or regularization by denoising (RED) framework [31].”
Polarimetric SAR Despeckling With Convolutional Neural Networks
Tucker, D., & Potter, L. C. (2022) · IEEE Transactions on Geoscience and Remote Sensing · doi:10.1109/tgrs.2022.3152068 · Cited by 27
“The MuLoG framework poses a maximum a posteriori (MAP) optimization problem that is solved by the plugand-play variant of the alternating direction method of multipliers (ADMM) [62], an approach which facilitates the use of single-channel AWGN denoising operators.”
Retinex Image Enhancement Based on Sequential Decomposition With a Plug-and-Play Framework
Wu, T., Wu, W., Yang, Y., Fan, F., & Zeng, T. (2023) · IEEE Transactions on Neural Networks and Learning Systems · doi:10.1109/tnnls.2023.3280037 · Cited by 36
“Earlier, the plug-and-play framework was proposed to solve the denoising problem [21], which allows the insertion of different denoisers for prior knowledge learning.”
Shi, B., & Liu, K. (2023) · IEEE/CAA Journal of Automatica Sinica · doi:10.1109/jas.2023.123543 · Cited by 27
Applications and Challenges of AI in PCB X-ray Inspection: A Comprehensive Study
Roy, A., Hasan, M. M. A., Ghosh, S., Varshney, N., Julia, J., Forghani, R., & Asadizanjani, N. (2025) · ACM Journal on Emerging Technologies in Computing Systems · doi:10.1145/3703457 · Cited by 29
“In this case, denoising algorithms are used as priors for model-based inversion [77, 83].”
Recovery Analysis for Plug-and-Play Priors using the Restricted Eigenvalue ConditionPreprint
Liu, J., Asif, M. S., Wohlberg, B. E., & Kamilov, U. S. (2021) · arXiv (Cornell University) · doi:10.48550/arxiv.2106.03668 · Cited by 50
“Plug-and-play priors (PnP) [10, 11] and regularization by denoising (RED) [12] are two methods related to CSGM that can also leverage pre-trained deep models as priors for inverse problems.”
Self-citationScientific Computational Imaging Code (SCICO)
Balke, T., Davis, F., García–Cardona, C., Majee, S., McCann, M. T., Pfister, L., & Wohlberg, B. E. (2022) · The Journal of Open Source Software · doi:10.21105/joss.04722 · Cited by 25
Self-citationPlug-and-Play Image Reconstruction Is a Convergent Regularization Method
Ebner, A., & Haltmeier, M. (2024) · IEEE Transactions on Image Processing · doi:10.1109/tip.2024.3361218 · Cited by 37
Deep Plug-and-Play Prior for Parallel MRI Reconstruction
Yazdanpanah, A. P., Afacan, O., & Warfield, S. K. (2019) · IEEE International Conference on Computer Vision Workshops · doi:10.1109/iccvw.2019.00489 · Cited by 25
“The plug-and-play prior framework [30] presented with an idea to utilize the denoiser without any regularization objective as proximal operator in an iterative method for image recovery.”
Distributed Iterative CT Reconstruction Using Multi-Agent Consensus Equilibrium
Sridhar, V., Wang, X., Buzzard, G. T., & Bouman, C. A. (2020) · IEEE Transactions on Computational Imaging · doi:10.1109/tci.2020.3008782 · Cited by 25
Self-citationCompressed Sensing-Based Robust Phase Retrieval via Deep Generative Priors
Shamshad, F., & Ahmed, A. M. (2020) · IEEE Sensors Journal · doi:10.1109/jsen.2020.3018751 · Cited by 25
“Plugand-play methods for phase retrieval include SPAR [29], BM3D-prGAMP [27], and plug-and-play ADMM [52], [53].”
Multi-Resolution Data Fusion for Super Resolution Imaging
Reid, E. J., Drummy, L. F., Bouman, C. A., & Buzzard, G. T. (2022) · IEEE Transactions on Computational Imaging · doi:10.1109/tci.2022.3140551 · Cited by 25
Self-citationVision-Language Gradient Descent-driven All-in-One Deep Unfolding Networks
Zeng, H., Wang, X., Chen, Y., Su, J., & Liu, J. (2025) · IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings - IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings · doi:10.1109/cvpr52734.2025.00705 · Cited by 24
Memory-Efficient Model-Based Deep Learning With Convergence and Robustness Guarantees
Pramanik, A., Zimmerman, M. B., & Jacob, M. (2023) · IEEE Transactions on Computational Imaging · doi:10.1109/tci.2023.3252268 · Cited by 29
“Plug and play methods replace the proximal mapping with off-the-shelf or CNN denoisersHθ [6], [13], [17], [33]:”
Plug-and-Play Split Gibbs Sampler: Embedding Deep Generative Priors in Bayesian Inference
Coeurdoux, F., Dobigeon, N., & Chainais, P. (2024) · IEEE Transactions on Image Processing · doi:10.1109/tip.2024.3404338 · Cited by 63
“To overcome these limitations, more and more deep learning based methods propose to combine DNN with conventional optimization algorithms within the so-called plug-and-play (PnP) framework [11].”
LINEARIZED ADMM AND FAST NONLOCAL DENOISING FOR EFFICIENT PLUG-AND-PLAY RESTORATION
Unni, V. S., Ghosh, S., & Chaudhury, K. N. (2018) · IEEE Global Conference on Signal and Information Processing · doi:10.1109/globalsip.2018.8646599 · Cited by 24
Effective new methods for automated parameter selection in regularized inverse problems
Sanders, T., Platte, R. B., & Skeel, R. D. (2020) · Applied Numerical Mathematics · doi:10.1016/j.apnum.2020.01.015 · Cited by 24
Exploiting prior knowledge about biological macromolecules in cryo-EM structure determination
Kimanius, D., Zickert, G., Nakane, T., Adler, J., Lunz, S., Schönlieb, C., Öktem, O., & Scheres, S. H. W. (2020) · IUCrJ · doi:10.1107/s2052252520014384 · Cited by 27
“Alternatively, algorithms based on ADMM and plug-and-play denoisers are of potential interest for investigation [44, 45].”
Armannsson, S. E., Ulfarsson, M. O., Sigurðsson, J., Nguyen, H. V., & Sveinsson, J. R. (2021) · Remote Sensing · doi:10.3390/rs13112192 · Cited by 24
“Here, regularization is performed with the Color Block Matching 3D Denoiser (C-BM3D) [39], used as a plug-and-play patch-based regularizer [40].”
Plug-and-Play Quantum Adaptive Denoiser for Deconvolving Poisson Noisy Images
Dutta, S., Basarab, A., Georgeot, B., & Kouamé, D. (2021) · IEEE Access · doi:10.1109/access.2021.3118608 · Cited by 26
“Since its initial development, the PnP scheme [25] is largely accepted for signal and image restoration problems due to its extremely promising performance [30]–[45].”
Restoration of speckle noise corrupted SAR images using regularization by denoising
Baraha, S., & Sahoo, A. K. (2022) · Journal of Visual Communication and Image Representation · doi:10.1016/j.jvcir.2022.103546 · Cited by 24
Approximation of Lipschitz Functions Using Deep Spline Neural Networks
Neumayer, S., Goujon, A., Bohra, P., & Unser, M. A. (2023) · SIAM Journal on Mathematics of Data Science · doi:10.1137/22m1504573 · Cited by 23
A Self-Supervised Deep Denoiser for Hyperspectral and Multispectral Image Fusion
Wang, Z., Ng, M. K., Michalski, J. R., & Zhuang, L. (2023) · IEEE Transactions on Geoscience and Remote Sensing · doi:10.1109/tgrs.2023.3303921 · Cited by 29
A New Recurrent Plug-and-Play Prior Based on the Multiple Self-Similarity Network
Song, G., Sun, Y., Liu, J., Wang, Z., & Kamilov, U. S. (2020) · IEEE Signal Processing Letters · doi:10.1109/lsp.2020.2977214 · Cited by 25
“We optimized the hyperparameters of TV and BM3D for SNR.”
InfluentialGradient-based discriminative modeling for blind image deblurring
Shao, W., Lin, Y., Liu, Y.-Y., Wang, L., Ge, Q., Bao, B., & Li, H. (2020) · Neurocomputing · doi:10.1016/j.neucom.2020.06.093 · Cited by 25
Venkatakrishnan, S. V., Mohan, K. A., Ziabari, A., & Bouman, C. A. (2021) · IEEE Signal Processing Magazine · doi:10.1109/msp.2021.3123594 · Cited by 23
“IV-A and then replacing certain blocks with trainable deep neural networks (DNN) in order to obtain high-quality reconstructions (including methods categorized as loop-unrolling algorithms [73] or plug-and-play priors [74], [75]) .”
Self-citationLearning Local Regularization for Variational Image Restoration
Prost, J., Houdard, A., Almansa, A., & Papadakis, N. (2021) · Lecture notes in computer science · doi:10.1007/978-3-030-75549-2_29 · Cited by 31
Data and Image Prior Integration for Image Reconstruction Using Consensus Equilibrium
Ghani, M. U., & Karl, W. C. (2021) · IEEE Transactions on Computational Imaging · doi:10.1109/tci.2021.3062986 · Cited by 22
“The plug-and-play framework (PnP-MBIR) [27] uses ADMM to split the original problem into sensor and imagedomain model sub-problems.”
InfluentialPellizzari, C. J., Spencer, M. F., & Bouman, C. A. (2020) · IEEE Transactions on Computational Imaging · doi:10.1109/tci.2020.3042948 · Cited by 22
Self-citationUnified Supervised-Unsupervised (SUPER) Learning for X-Ray CT Image Reconstruction
Ye, S., Li, Z., McCann, M. T., Long, Y. Q., & Ravishankar, S. (2021) · IEEE Transactions on Medical Imaging · doi:10.1109/tmi.2021.3095310 · Cited by 22
“play [49], with which various unsupervised and supervised regularizers can be plugged.”
Gelvez, T., Argüello, H., & Foi, A. (2022) · IEEE Transactions on Geoscience and Remote Sensing · doi:10.1109/tgrs.2022.3203294 · Cited by 22
“69 The low-rank priors are combined in the problem through 70 an implicit regularization function, which is solved numer- 71 ically by the versatile and flexible plug-and-play with 72 alternating direction method of multipliers (PnP-ADMM) 73 framework [23], [24], [25], [26], [27].”
RISING: A new framework for model-based few-view CT image reconstruction with deep learning
Evangelista, D., Morotti, E., & Piccolomini, E. L. (2022) · Computerized Medical Imaging and Graphics · doi:10.1016/j.compmedimag.2022.102156 · Cited by 22
Provable Preconditioned Plug-and-Play Approach for Compressed Sensing MRI Reconstruction
Hong, T., Xu, X., Hu, J., & Fessler, J. A. (2024) · IEEE Transactions on Computational Imaging · doi:10.1109/tci.2024.3477329 · Cited by 22
“Plug-and-Play (PnP) [21] is an alternative to DL that leverages the most effective image denoisers, such as BM3D [22] or DnCNN [23], leading to state-of-the-art performance in various imaging tasks [24–29].”
Iterative Shrinkage Algorithm for Patch-Smoothness Regularized Medical Image Recovery
Mohsin, Y. Q., Ongie, G., & Jacob, M. (2015) · IEEE Transactions on Medical Imaging · doi:10.1109/tmi.2015.2398466 · Cited by 24
Wasserstein-Based Projections with Applications to Inverse Problems
Heaton, H., Fung, S. W., Lin, A. T., Osher, S., & Yin, W. (2022) · SIAM Journal on Mathematics of Data Science · doi:10.1137/20m1376790 · Cited by 28
“A generalization of using data-driven regularization consists of Plug-and-Play (PnP) methods [25, 28, 94], which replace the proximal operators in an optimization algorithm with data-driven operators.”
Wasserstein Patch Prior for Image Superresolution
Hertrich, J., Houdard, A., & Redenbach, C. (2022) · IEEE Transactions on Computational Imaging · doi:10.1109/tci.2022.3199600 · Cited by 30
A Fast Stochastic Plug-and-Play ADMM for Imaging Inverse ProblemsPreprint
Tang, J., & Davies, M. E. (2020) · arXiv (Cornell University) · doi:10.48550/arxiv.2006.11630 · Cited by 22
“In this work we study and propose a stochastic gradient extension of the PnP-ADMM algorithm 1 [1]: where in each iteration an exact proximal step on the data-fidelity term f ( x ) is computed with a constant step-size τ : Algorithm 1 — Stochastic PnP-ADMM In classical ADMM the step-size τ can be any…”
Computational Imaging Through Atmospheric Turbulence
Chan, S. H., & Chimitt, N. (2023) · Foundations and Trends® in Computer Graphics and Vision · doi:10.1561/0600000103 · Cited by 28
Liaudat, T. I., Mars, M., Price, M. A., Pereyra, M., Betcke, M. M., & McEwen, J. D. (2024) · RAS Techniques and Instruments · doi:10.1093/rasti/rzae030 · Cited by 27
A Decade Review of Video Compressive Sensing: A Roadmap to Practical Applications
Zhang, Z., Zheng, S., Qiu, M., Situ, G., Brady, D. J., Dai, Q., Suo, J., & Yuan, X. (2024) · Engineering · doi:10.1016/j.eng.2024.08.013 · Cited by 21
Discriminative Transfer Learning for General Image Restoration
Xiao, L., Heide, F., Heidrich, W., Schölkopf, B., & Hirsch, M. (2018) · IEEE Transactions on Image Processing · doi:10.1109/tip.2018.2831925 · Cited by 22
“12 by alternately solving the following 3 subproblems: where prox Θ is from our previous training, and the v t sub-problem is approximated by running BM3D software on x t − 1 with noise parameter ts following [33, 14].”
On the Proof of Fixed-Point Convergence for Plug-and-Play ADMM
Gavaskar, R. G., & Chaudhury, K. N. (2019) · IEEE Signal Processing Letters · doi:10.1109/lsp.2019.2950611 · Cited by 21
BP-DIP: A Backprojection based Deep Image Prior
Zukerman, J., Tirer, T., & Giryes, R. (2020) · European Signal Processing Conference (EUSIPCO) · doi:10.23919/eusipco47968.2020.9287540 · Cited by 21
“Another recent variant of [19] modifies the way that the denoising engine is used to regularize the inverse problem [20], [21].”
An image reconstruction framework for characterizing initial visual encoding
Zhang, L.-Q., Cottaris, N. P., & Brainard, D. H. (2022) · eLife · doi:10.7554/elife.71132 · Cited by 20
Scalable precision wide-field imaging in radio interferometry – II. AIRI validated on ASKAP data
Wilber, A., Dabbech, A., Terris, M., Jackson, A., & Wiaux, Y. (2023) · Monthly Notices of the Royal Astronomical Society · doi:10.1093/mnras/stad1353 · Cited by 20
SoDaCam: Software-defined Cameras via Single-Photon Imaging
Sundar, V., Ardelean, A., Swedish, T. B., Bruschini, C. E., Charbon, E., & Gupta, M. (2023) · IEEE/CVF International Conference on Computer Vision (ICCV) · doi:10.1109/iccv51070.2023.00750 · Cited by 27
“Related Work Reinterpretable imaging has previously been explored at the granularity of light rays [3], by modulating the plenoptic function, and at the level of spatio-temporal voxels [20], by using fast per-pixel shutters.”
Iterative reconstruction for limited-angle CT using implicit neural representation
Lee, J., & Baek, J. (2024) · Physics in Medicine and Biology · doi:10.1088/1361-6560/ad3c8e · Cited by 20
Plug-and-Play Regularization on Magnitude With Deep Priors for 3D Near-Field MIMO Imaging
Oral, O., & Öktem, F. S. (2024) · IEEE Transactions on Computational Imaging · doi:10.1109/tci.2024.3396388 · Cited by 20
“Learned Plug-and-Play (PnP) regularization [26], [29], [31], [32], [33] and unrolling-based methods [28], [30], [34] are examples of such approaches.”
Plug-and-Play Algorithms for Dynamic Non-line-of-sight Imaging
Ye, J., Hong, Y., Su, X., Yuan, X., & Xu, F. (2024) · ACM Transactions on Graphics · doi:10.1145/3665139 · Cited by 20
Convergent Regularization in Inverse Problems and Linear Plug-and-Play Denoisers
Hauptmann, A., Mukherjee, S., Schönlieb, C., & Sherry, F. (2024) · Foundations of Computational Mathematics · doi:10.1007/s10208-024-09654-x · Cited by 20
Learning Convex Regularizers for Optimal Bayesian Denoising
Nguyen, H. Q., Bostan, E., & Unser, M. A. (2017) · IEEE Transactions on Signal Processing · doi:10.1109/tsp.2017.2777407 · Cited by 20
“The main reason of the momentum is that Bayesian estimators can be directly integrated—as “denoisers”—into algorithms that are designed for more sophisticated inverse problems [4].”
Metadata
- Authors
- Venkatakrishnan, S. V., Bouman, C. A., & Wohlberg, B.
- Affiliations
- Purdue University West Lafayette (US)
- Los Alamos National Laboratory (US)
- Venue
- 2013 IEEE Global Conference on Signal and Information Processing
- Publisher
- IEEE
- Published
- 2013-12
- Pages
- 945-948
- ISSN
- 2376-4066 (ISSN-L)
- OpenAlex
- W2087416986
- Cited by
- 1,178
- References
- 32
- Citations by year
- 2026: 99 · 2025: 172 · 2024: 147 · 2023: 177 · 2022: 121 · 2021: 182 · 2020: 102 · 2019: 75 · 2018: 52 · 2017: 24 · 2016: 16 · 2015: 8 · 2014: 2
- FWCI
- 5.53
- Type
- conference paper
- Language
- en
- Open access
- closed
- Locations
- IEEE Global Conference on Signal and Information Processing (publishedVersion)
- Purdue e-Pubs (Purdue University System) (submittedVersion)
- Topics
- Physical Sciences › Computer Science › Computer Vision and Pattern Recognition › Image and Signal Denoising Methods
- Physical Sciences › Mathematics › Mathematical Physics › Numerical methods in inverse problems
- Health Sciences › Medicine › Radiology, Nuclear Medicine and Imaging › Medical Imaging Techniques and Applications
- Keywords
- plug-and-play priors, model-based reconstruction, image denoising, imaging inverse problems, X-ray CT, electron microscopy, magnetic resonance imaging, ultrasonography, prior models, software integration
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
Data sources
Bibliographic data: Semantic Scholar, OpenAlex, Crossref.
Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).
Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.