21,254 Hits in 5.5 sec

Compressed Sensing with Deep Image Prior and Learned Regularization [article]

Dave Van Veen, Ajil Jalal, Mahdi Soltanolkotabi, Eric Price, Sriram Vishwanath, Alexandros G. Dimakis
2020 arXiv   pre-print
We propose a novel method for compressed sensing recovery using untrained deep generative models.  ...  Our method is based on the recently proposed Deep Image Prior (DIP), wherein the convolutional weights of the network are optimized to match the observed measurements.  ...  Compressed Sensing with Deep Image Prior (CS-DIP) Our approach is to find a set of weights for the convolutional network such that the measurement matrix applied to the network output, i.e.  ... 
arXiv:1806.06438v4 fatcat:hqjraly4vrhwbnhpjvsci2fosa

MRI Reconstruction Using Deep Bayesian Inference [article]

GuanXiong Luo, Na Zhao, Wenhao Jiang, Peng Cao
2019 arXiv   pre-print
The proposed method generally achieved more than 5 dB peak signal-to-noise ratio improvement for compressed sensing and parallel imaging reconstructions compared with the other methods.  ...  Results: The proposed method showed an improved performance in preserving image details and reducing aliasing artifacts, compared with GRAPPA, ℓ_1-ESPRiT, and MODL, a state-of-the-art deep learning reconstruction  ...  in the compressed sensing, parallel imaging, and deep learning reconstructions.  ... 
arXiv:1909.01127v1 fatcat:sojleyhhkrc5rg2aam55owxkpe

DNU: Deep Non-Local Unrolling for Computational Spectral Imaging

Lizhi Wang, Chen Sun, Maoqing Zhang, Ying Fu, Hua Huang
2020 2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)  
Our data-driven prior is integrated as a regularizer into the reconstruction problem. Then, we propose to unroll the reconstruction problem into an optimization-inspired deep neural network.  ...  First, we introduce a novel data-driven prior that can adaptively exploit both the local and non-local correlations among the spectral image.  ...  Recently, by leveraging the large datasets, the concept of deep denoising prior has been proposed by learning an implicit but more accurate prior based on deep learning [61, 43, 37] .  ... 
doi:10.1109/cvpr42600.2020.00173 dblp:conf/cvpr/WangSZF020 fatcat:jhzpiwj2b5fjjd6hotf3gclxga

Review Article: Model Meets Deep Learning in Image Inverse Problems

Na Wang & Jian Sun
2020 CSIAM Transactions on Applied Mathematics  
In recent years, deep learning has been introduced to image inverse problems by learning to invert image sensing or degradation process.  ...  But these methods require good design of image prior or regularizer that is hard to be hand-crafted.  ...  Acknowledgments This work was supported by NSFC (11971373, 11690011, U1811461, 61721002) and National Key R&D Program 2018AAA0102201.  ... 
doi:10.4208/csiam-am.2020-0016 fatcat:peaina2vorg23ow5seswsi7pzu

Invertible generative models for inverse problems: mitigating representation error and dataset bias [article]

Muhammad Asim, Max Daniels, Oscar Leong, Ali Ahmed, Paul Hand
2020 arXiv   pre-print
We additionally compare performance for compressive sensing to unlearned methods, such as the deep decoder, and we establish theoretical bounds on expected recovery error in the case of a linear invertible  ...  sensing, and inpainting.  ...  The Figure 7 . 7 Compressive sensing on FFHQ images with m = 7, 500 (≈ 20%) of measurements. Visual comparisons: CS under the Glow prior, PGGAN prior, and the overparameterized Deep Decoder prior.  ... 
arXiv:1905.11672v4 fatcat:hgpfoh6frfa4thyxvhmqjzqomi

2019 Index IEEE Transactions on Computational Imaging Vol. 5

2019 IEEE Transactions on Computational Imaging  
., +, TCI Sept. 2019 344-353 Compressed sensing Monotone FISTA With Variable Acceleration for Compressed Sensing Mag- netic Resonance Imaging.  ...  Lin, C.Y., +, TCI March 2019 17-26 Data compression Solving Inverse Computational Imaging Problems Using Deep Pixel-Level Prior.  ...  Surgery Multiresolution Cube Propagation for 3-D Ultrasound Image Reconstruction. Dong  ... 
doi:10.1109/tci.2019.2959176 fatcat:g7nuyesverg2xbjwbzuyp6ovyy

Compressive Spectral Image Reconstruction using Deep Prior and Low-Rank Tensor Representation [article]

Jorge Bacca, Yesid Fonseca, Henry Arguello
2021 arXiv   pre-print
In general, the reconstruction methods are based on hand-crafted priors used as regularizers in optimization algorithms or recent deep neural networks employed as an image generator to learn a non-linear  ...  The proposed scheme is obtained by minimizing the ℓ_2-norm distance between the compressive measurements and the predicted measurements, and the desired recovered spectral image is formed just before the  ...  and (·) : R × × → R is a regularization function that imposes particular image priors with as the regularization parameter [36] .  ... 
arXiv:2101.07424v2 fatcat:ajcwarf7inbznb7vtzhdjpgsli

JR2net: A Joint Non-Linear Representation and Recovery Network for Compressive Spectral Imaging [article]

Brayan Monroy, Jorge Bacca, Henry Arguello
2022 arXiv   pre-print
Deep learning models are state-of-the-art in compressive spectral imaging (CSI) recovery.  ...  These methods use a deep neural network (DNN) as an image generator to learn non-linear mapping from compressed measurements to the spectral image.  ...  Model-based Optimization with Deep Priors These methods are based on iterative techniques that replace the hand-crafted prior with a deep neural network (DNN) used to learn a deep prior of the spectral  ... 
arXiv:2205.07770v1 fatcat:sajbzktaz5akjihfrlfxlrq7se

2020 Index IEEE Transactions on Computational Imaging Vol. 6

2020 IEEE Transactions on Computational Imaging  
., +, TCI 2020 419-433 MRI Super-Resolution With Ensemble Learning and Complementary Priors.  ...  ., +, TCI 2020 503-517 Automated Regularization Parameter Selection Using Continuation Based Proximal Method for Compressed Sensing MRI.  ... 
doi:10.1109/tci.2021.3054596 fatcat:puij7ztll5ai7alxrmqzsupcny

Compressive sensing with un-trained neural networks: Gradient descent finds a smooth approximation

Reinhard Heckel, Mahdi Soltanolkotabi
2020 International Conference on Machine Learning  
They are capable of solving standard inverse problems such as denoising and compressive sensing with excellent results by simply fitting a neural network model to measurements from a single image or signal  ...  a natural image from few measurements when trained with gradient descent until convergence.  ...  ., 2018) and compressive sensing, no further regularization is necessary!  ... 
dblp:conf/icml/HeckelS20 fatcat:idg3mmzwejahhpho7hwwpulopy

A Compressed Reconstruction Network Combining Deep Image Prior and Autoencoding Priors for Single-Pixel Imaging

Jian Lin, Qiurong Yan, Shang Lu, Yongjian Zheng, Shida Sun, Zhen Wei
2022 Photonics  
Single-pixel imaging (SPI) is a promising imaging scheme based on compressive sensing.  ...  DLCNets learn prior distributions of real pictures from massive datasets, while the Deep Image Prior (DIP) uses a neural network′s own structural prior to solve inverse problems without requiring a lot  ...  Deep-Learning-based Compressed Sensing Reconstruction Network Traditional compressive reconstruction algorithm: In single-pixel imaging, the number of measurements is much smaller than the number of image  ... 
doi:10.3390/photonics9050343 fatcat:unyy4gik2fgk3frgql3tmqtvjm

A review on deep learning MRI reconstruction without fully sampled k-space

Gushan Zeng, Yi Guo, Jiaying Zhan, Zi Wang, Zongying Lai, Xiaofeng Du, Xiaobo Qu, Di Guo
2021 BMC Medical Imaging  
Compressed sensing and parallel imaging are two common techniques to accelerate MRI reconstruction.  ...  Furthermore, we may anticipate that the combination of traditional methods and deep learning will produce better reconstruction results.  ...  Acknowledgements The authors thank Zhangren Tu, Chen Qian, and Haoming Fang for valuable discussions.  ... 
doi:10.1186/s12880-021-00727-9 pmid:34952572 pmcid:PMC8710001 fatcat:q3t6htimkbcj5gvfd3qbi6ssem

Computational MRI: Compressive Sensing and Beyond [From the Guest Editors]

Mathews Jacob, Jong Chul Ye, Leslie Ying, Mariya Doneva
2020 IEEE Signal Processing Magazine  
sensing, and seismic imaging.  ...  Advances in computational MRI were primarily driven in the last decade by parallel image acquisition using multiple receiver coils and compressed sensing (CS).  ...  Namrata Vaswani, Clem Karl, SPM's Editor-in-Chief Robert Heath, and SPM's entire editorial board for their helpful feedback and suggestions for the content of this special issue.  ... 
doi:10.1109/msp.2019.2953993 pmid:33623355 pmcid:PMC7899158 fatcat:kqogwy52r5grbezsfjp2mzxs6e

Algorithmic Guarantees for Inverse Imaging with Untrained Network Priors [article]

Gauri Jagatap, Chinmay Hegde
2020 arXiv   pre-print
Deep neural networks as image priors have been recently introduced for problems such as denoising, super-resolution and inpainting with promising performance gains over hand-crafted image priors such as  ...  We also show both theoretically as well as empirically that with deep network priors, one can achieve better compression rates for the same image quality compared to hand crafted priors.  ...  We introduce a novel formulation, to solve compressive phase retrieval with fewer measurements as compared to state-of-art.  ... 
arXiv:1906.08763v2 fatcat:lntgt2cqh5d35eayht5mah467e

Learning Nonlocal Sparse and Low-Rank Models for Image Compressive Sensing [article]

Zhiyuan Zha, Bihan Wen, Xin Yuan, Saiprasad Ravishankar, Jiantao Zhou, Ce Zhu
2022 arXiv   pre-print
Such methods can effectively model the structures of image patches by optimizing their sparse representations or learning deep neural networks, while preserving the known or modeled sensing process.  ...  This article reviews some recent works in image CS tasks with a focus on the advanced GSR and LR based methods.  ...  The most recent trend combined deep neural networks with NSS prior for image CS, such as learning nonlocal regularized compressed sensing network (NLR-CSNet) [34] , and image CS framework using nonlocal  ... 
arXiv:2203.09656v3 fatcat:zkmkewwbcbhctj4cpsibthreru
« Previous Showing results 1 — 15 out of 21,254 results