Deep learning for tomographic image reconstruction

G Wang, JC Ye, B De Man - Nature machine intelligence, 2020‏ - nature.com
Deep-learning-based tomographic imaging is an important application of artificial
intelligence and a new frontier of machine learning. Deep learning has been widely used in …

AI-based reconstruction for fast MRI—A systematic review and meta-analysis

Y Chen, CB Schönlieb, P Liò, T Leiner… - Proceedings of the …, 2022‏ - ieeexplore.ieee.org
Compressed sensing (CS) has been playing a key role in accelerating the magnetic
resonance imaging (MRI) acquisition process. With the resurgence of artificial intelligence …

Adaptive diffusion priors for accelerated MRI reconstruction

A Güngör, SUH Dar, Ş Öztürk, Y Korkmaz… - Medical image …, 2023‏ - Elsevier
Deep MRI reconstruction is commonly performed with conditional models that de-alias
undersampled acquisitions to recover images consistent with fully-sampled data. Since …

Plug-and-play methods for integrating physical and learned models in computational imaging: Theory, algorithms, and applications

US Kamilov, CA Bouman, GT Buzzard… - IEEE Signal …, 2023‏ - ieeexplore.ieee.org
Plug-and-play (PnP) priors constitute one of the most widely used frameworks for solving
computational imaging problems through the integration of physical models and learned …

Unsupervised MRI reconstruction via zero-shot learned adversarial transformers

Y Korkmaz, SUH Dar, M Yurt, M Özbey… - IEEE Transactions on …, 2022‏ - ieeexplore.ieee.org
Supervised reconstruction models are characteristically trained on matched pairs of
undersampled and fully-sampled data to capture an MRI prior, along with supervision …

Deep equilibrium architectures for inverse problems in imaging

D Gilton, G Ongie, R Willett - IEEE Transactions on …, 2021‏ - ieeexplore.ieee.org
Recent efforts on solving inverse problems in imaging via deep neural networks use
architectures inspired by a fixed number of iterations of an optimization method. The number …

Self‐supervised learning of physics‐guided reconstruction neural networks without fully sampled reference data

B Yaman, SAH Hosseini, S Moeller… - Magnetic resonance …, 2020‏ - Wiley Online Library
Purpose To develop a strategy for training a physics‐guided MRI reconstruction neural
network without a database of fully sampled data sets. Methods Self‐supervised learning via …

Flot: Scene flow on point clouds guided by optimal transport

G Puy, A Boulch, R Marlet - European conference on computer vision, 2020‏ - Springer
We propose and study a method called FLOT that estimates scene flow on point clouds. We
start the design of FLOT by noticing that scene flow estimation on point clouds reduces to …

Self-supervised neural networks for spectral snapshot compressive imaging

Z Meng, Z Yu, K Xu, X Yuan - Proceedings of the IEEE/CVF …, 2021‏ - openaccess.thecvf.com
We consider using untrained neural networks to solve the reconstruction problem of
snapshot compressive imaging (SCI), which uses a two-dimensional (2D) detector to …

Gradient step denoiser for convergent plug-and-play

S Hurault, A Leclaire, N Papadakis - arxiv preprint arxiv:2110.03220, 2021‏ - arxiv.org
Plug-and-Play methods constitute a class of iterative algorithms for imaging problems where
regularization is performed by an off-the-shelf denoiser. Although Plug-and-Play methods …