no code implementations • 23 Feb 2024 • Sébastien Herbreteau, Charles Kervrann
Image denoising is probably the oldest and still one of the most active research topic in image processing.
no code implementations • 21 Feb 2024 • Sébastien Herbreteau, Charles Kervrann
We propose a unified view of non-local methods for single-image denoising, for which BM3D is the most popular representative, that operate by gathering noisy patches together according to their similarities in order to process them collaboratively.
1 code implementation • NeurIPS 2023 • Sébastien Herbreteau, Emmanuel Moebel, Charles Kervrann
In many information processing systems, it may be desirable to ensure that any change of the input, whether by shifting or scaling, results in a corresponding change in the system response.
1 code implementation • 13 Mar 2023 • Lisa Balsollier, Frédéric Lavancier, Jean Salamero, Charles Kervrann
Generators of space-time dynamics in bioimaging have become essential to build ground truth datasets for image processing algorithm evaluation such as biomolecule detectors and trackers, as well as to generate training datasets for deep learning algorithms.
1 code implementation • 1 Dec 2022 • Sébastien Herbreteau, Charles Kervrann
We introduce a parametric view of non-local two-step denoisers, for which BM3D is a major representative, where quadratic risk minimization is leveraged for unsupervised optimization.
1 code implementation • 1 Mar 2022 • Sébastien Herbreteau, Charles Kervrann
We propose a unified view of unsupervised non-local methods for image denoising that linearily combine noisy image patches.
1 code implementation • 31 Jul 2021 • Sébastien Herbreteau, Charles Kervrann
This work tackles the issue of noise removal from images, focusing on the well-known DCT image denoising algorithm.
no code implementations • NeurIPS 2014 • Charles Kervrann
Patch-based methods have been widely used for noise reduction in recent years.
no code implementations • 22 Jul 2014 • Denis Fortun, Patrick Bouthemy, Charles Kervrann
The idea is to supply local motion candidates at every pixel in a first step, and then to combine them to determine the global optical flow field in a second step.