State-of-the-art approaches for image deconvolution problems, including modern deep learning architectures
M Makarkin, D Bratashov - Micromachines, 2021 - mdpi.com
In modern digital microscopy, deconvolution methods are widely used to eliminate a number
of image defects and increase resolution. In this review, we have divided these methods into …
of image defects and increase resolution. In this review, we have divided these methods into …
Single image defocus deblurring via implicit neural inverse kernels
Single image defocus deblurring (SIDD) is a challenging task due to the spatially-varying
nature of defocus blur, characterized by per-pixel point spread functions (PSFs). Existing …
nature of defocus blur, characterized by per-pixel point spread functions (PSFs). Existing …
Image denoising in the deep learning era
S Izadi, D Sutton, G Hamarneh - Artificial Intelligence Review, 2023 - Springer
Over the last decade, the number of digital images captured per day has increased
exponentially, due to the accessibility of imaging devices. The visual quality of photographs …
exponentially, due to the accessibility of imaging devices. The visual quality of photographs …
IDENet: Implicit degradation estimation network for efficient blind super resolution
AH Khan, C Micheloni… - Proceedings of the IEEE …, 2024 - openaccess.thecvf.com
Blind image super-resolution (SR) aims to recover high-resolution (HR) images from low-
resolution (LR) inputs hindered by unknown degradation. Existing blind SR methods exploit …
resolution (LR) inputs hindered by unknown degradation. Existing blind SR methods exploit …
Learning deep non-blind image deconvolution without ground truths
Non-blind image deconvolution (NBID) is about restoring a latent sharp image from a
blurred one, given an associated blur kernel. Most existing deep neural networks for NBID …
blurred one, given an associated blur kernel. Most existing deep neural networks for NBID …
Feather-light Fourier domain adaptation in magnetic resonance imaging
Generalizability of deep learning models may be severely affected by the difference in the
distributions of the train (source domain) and the test (target domain) sets, eg, when the sets …
distributions of the train (source domain) and the test (target domain) sets, eg, when the sets …
Clean implicit 3d structure from noisy 2d stem images
H Kniesel, T Ropinski, T Bergner… - Proceedings of the …, 2022 - openaccess.thecvf.com
Abstract Scanning Transmission Electron Microscopes (STEMs) acquire 2D images of a 3D
sample on the scale of individual cell components. Unfortunately, these 2D images can be …
sample on the scale of individual cell components. Unfortunately, these 2D images can be …
Autofocusing+: noise-resilient motion correction in magnetic resonance imaging
Image corruption by motion artifacts is an ingrained problem in Magnetic Resonance
Imaging (MRI). In this work, we propose a neural network-based regularization term to …
Imaging (MRI). In this work, we propose a neural network-based regularization term to …
[PDF][PDF] The use of the Kolmogorov-Wiener filter for prediction of heavy-tail stationary processes.
V Gorev, A Gusev, V Korniienko - IntelITSIS, 2022 - ceur-ws.org
We investigate the possibility of the practical use of the Kolmogorov–Wiener filter for the
prediction of a heavy-tail stationary random process. A discrete process and a discrete filter …
prediction of a heavy-tail stationary random process. A discrete process and a discrete filter …
Lightweight Prompt Learning Implicit Degradation Estimation Network for Blind Super Resolution
Blind image super-resolution (SR) aims to recover a high-resolution (HR) image from its low-
resolution (LR) counterpart under the assumption of unknown degradations. Many existing …
resolution (LR) counterpart under the assumption of unknown degradations. Many existing …