Search Results for author: Jean-Luc Starck

Found 28 papers, 19 papers with code

Hybrid Physical-Neural ODEs for Fast N-body Simulations

1 code implementation12 Jul 2022 Denise Lanzieri, François Lanusse, Jean-Luc Starck

We present a new scheme to compensate for the small-scales approximations resulting from Particle-Mesh (PM) schemes for cosmological N-body simulations.

Rethinking data-driven point spread function modeling with a differentiable optical model

1 code implementation9 Mar 2022 Tobias Liaudat, Jean-Luc Starck, Martin Kilbinger, Pierre-Antoine Frugier

Even though observations of the PSF are available at some positions of the field of view (FOV), they are undersampled, noisy, and integrated in wavelength in the instrument's passband.

Astronomy Super-Resolution

Probabilistic Mass Mapping with Neural Score Estimation

no code implementations14 Jan 2022 Benjamin Remy, Francois Lanusse, Niall Jeffrey, Jia Liu, Jean-Luc Starck, Ken Osato, Tim Schrabback

We introduce a novel methodology allowing for efficient sampling of the high-dimensional Bayesian posterior of the weak lensing mass-mapping problem, and relying on simulations for defining a fully non-Gaussian prior.

Stable Long-Term Recurrent Video Super-Resolution

1 code implementation CVPR 2022 Benjamin Naoto Chiche, Arnaud Woiselle, Joana Frontera-Pons, Jean-Luc Starck

Finally, we introduce a new framework of recurrent VSR networks that is both stable and competitive, based on Lipschitz stability theory.

Video Super-Resolution

Rethinking the modeling of the instrumental response of telescopes with a differentiable optical model

1 code implementation24 Nov 2021 Tobias Liaudat, Jean-Luc Starck, Martin Kilbinger, Pierre-Antoine Frugier

By adding a differentiable optical forward model into the modeling framework, we change the data-driven modeling space from the pixels to the wavefront.

Super-Resolution

Is good old GRAPPA dead?

2 code implementations1 Jun 2021 Zaccharie Ramzi, Alexandre Vignaud, Jean-Luc Starck, Philippe Ciuciu

We perform a qualitative analysis of performance of XPDNet, a state-of-the-art deep learning approach for MRI reconstruction, compared to GRAPPA, a classical approach.

MRI Reconstruction

SHINE: SHaring the INverse Estimate from the forward pass for bi-level optimization and implicit models

2 code implementations ICLR 2022 Zaccharie Ramzi, Florian Mannel, Shaojie Bai, Jean-Luc Starck, Philippe Ciuciu, Thomas Moreau

In Deep Equilibrium Models (DEQs), the training is performed as a bi-level problem, and its computational complexity is partially driven by the iterative inversion of a huge Jacobian matrix.

Hyperparameter Optimization

Deep Unrolled Network for Video Super-Resolution

no code implementations23 Feb 2021 Benjamin Naoto Chiche, Arnaud Woiselle, Joana Frontera-Pons, Jean-Luc Starck

Yet, they fail to incorporate some knowledge about the image formation model, which limits their flexibility.

Image Restoration Video Super-Resolution

Starlet l1-norm for weak lensing cosmology

no code implementations5 Jan 2021 Virginia Ajani, Jean-Luc Starck, Valeria Pettorino

We present a new summary statistic for weak lensing observables, higher than second order, suitable for extracting non-Gaussian cosmological information and inferring cosmological parameters.

Cosmology and Nongalactic Astrophysics

Density Compensated Unrolled Networks for Non-Cartesian MRI Reconstruction

1 code implementation5 Jan 2021 Zaccharie Ramzi, Jean-Luc Starck, Philippe Ciuciu

Deep neural networks have recently been thoroughly investigated as a powerful tool for MRI reconstruction.

MRI Reconstruction

Denoising Score-Matching for Uncertainty Quantification in Inverse Problems

1 code implementation16 Nov 2020 Zaccharie Ramzi, Benjamin Remy, Francois Lanusse, Jean-Luc Starck, Philippe Ciuciu

Deep neural networks have proven extremely efficient at solving a wide rangeof inverse problems, but most often the uncertainty on the solution they provideis hard to quantify.

Denoising MRI Reconstruction

XPDNet for MRI Reconstruction: an application to the 2020 fastMRI challenge

2 code implementations15 Oct 2020 Zaccharie Ramzi, Philippe Ciuciu, Jean-Luc Starck

We present a new neural network, the XPDNet, for MRI reconstruction from periodically under-sampled multi-coil data.

MRI Reconstruction

Semi-supervised dictionary learning with graph regularization and active points

2 code implementations13 Sep 2020 Khanh-Hung Tran, Fred-Maurice Ngole-Mboula, Jean-Luc Starck, Vincent Prost

Supervised Dictionary Learning has gained much interest in the recent decade and has shown significant performance improvements in image classification.

Dictionary Learning General Classification +1

Manifold attack

1 code implementation13 Sep 2020 Khanh-Hung Tran, Fred-Maurice Ngole-Mboula, Jean-Luc Starck

Machine Learning in general and Deep Learning in particular has gained much interest in the recent decade and has shown significant performance improvements for many Computer Vision or Natural Language Processing tasks.

Natural Language Processing

Deep Learning for space-variant deconvolution in galaxy surveys

no code implementations1 Nov 2019 Florent Sureau, Alexis Lechat, Jean-Luc Starck

Deconvolution of large survey images with millions of galaxies requires to develop a new generation of methods which can take into account a space variant Point Spread Function (PSF) and have to be at the same time accurate and fast.

Image Reconstruction

Deep learning dark matter map reconstructions from DES SV weak lensing data

2 code implementations1 Aug 2019 Niall Jeffrey, François Lanusse, Ofer Lahav, Jean-Luc Starck

With a validation set of 8000 simulated DES SV data realisations, compared to Wiener filtering with a fixed power spectrum, the DeepMass method improved the mean-square-error (MSE) by 11 per cent.

Cosmology and Nongalactic Astrophysics

Semi-supervised dual graph regularized dictionary learning

no code implementations11 Dec 2018 Khanh-Hung Tran, Fred-Maurice Ngole-Mboula, Jean-Luc Starck

In this paper, we propose a semi-supervised dictionary learning method that uses both the information in labelled and unlabelled data and jointly trains a linear classifier embedded on the sparse codes.

Dictionary Learning

Distinguishing standard and modified gravity cosmologies with machine learning

no code implementations25 Oct 2018 Austin Peel, Florian Lalande, Jean-Luc Starck, Valeria Pettorino, Julian Merten, Carlo Giocoli, Massimo Meneghetti, Marco Baldi

We present a convolutional neural network to identify distinct cosmological scenarios based on the weak-lensing maps they produce.

Cosmology and Nongalactic Astrophysics

On the dissection of degenerate cosmologies with machine learning

no code implementations25 Oct 2018 Julian Merten, Carlo Giocoli, Marco Baldi, Massimo Meneghetti, Austin Peel, Florian Lalande, Jean-Luc Starck, Valeria Pettorino

Based on the DUSTGRAIN-pathfinder suite of simulations, we investigate observational degeneracies between nine models of modified gravity and massive neutrinos.

BIG-bench Machine Learning General Classification +1

Space variant deconvolution of galaxy survey images

2 code implementations7 Mar 2017 Samuel Farrens, Jean-Luc Starck, Fred Maurice Ngolè Mboula

This work introduces the use of the low-rank matrix approximation as a regularisation prior for galaxy image deconvolution and compares its performance with a standard sparse regularisation technique.

Instrumentation and Methods for Astrophysics Cosmology and Nongalactic Astrophysics

Sparsity and adaptivity for the blind separation of partially correlated sources

1 code implementation9 Dec 2014 Jerome Bobin, Jeremy Rapin, Anthony Larue, Jean-Luc Starck

Blind source separation (BSS) is a very popular technique to analyze multichannel data.

Super-resolution method using sparse regularization for point-spread function recovery

no code implementations16 Oct 2014 Fred Maurice Ngolè Mboula, Jean-Luc Starck, Samuel Ronayette, Koryo Okumura, Jérôme Amiaux

In large-scale spatial surveys, such as the forthcoming ESA Euclid mission, images may be undersampled due to the optical sensors sizes.

Super-Resolution

NMF with Sparse Regularizations in Transformed Domains

1 code implementation29 Jul 2014 Jérémy Rapin, Jérôme Bobin, Anthony Larue, Jean-Luc Starck

In this article, we show how a sparse NMF algorithm coined non-negative generalized morphological component analysis (nGMCA) can be extended to impose non-negativity in the direct domain along with sparsity in a transformed domain, with both analysis and synthesis formulations.

Sparse and Non-Negative BSS for Noisy Data

1 code implementation26 Aug 2013 Jérémy Rapin, Jérôme Bobin, Anthony Larue, Jean-Luc Starck

In this context, it is fundamental that the sources to be estimated present some diversity in order to be efficiently retrieved.

Astronomical Image Denoising Using Dictionary Learning

no code implementations12 Apr 2013 Simon Beckouche, Jean-Luc Starck, Jalal Fadili

Astronomical images suffer a constant presence of multiple defects that are consequences of the intrinsic properties of the acquisition equipments, and atmospheric conditions.

Dictionary Learning Image Denoising

Cannot find the paper you are looking for? You can Submit a new open access paper.