9 papers
Principled MAP estimation for inverse problems: bridging the gap between convergence and performance
Alexandre Lagier, Valentine Tosel, Anne Gagneux +2
Pretrained denoisers provide a powerful way to incorporate image priors into restoration algorithms. Plug-and-Play and RED approaches exploit fixed-noise-level denoisers within fir…
Beyond the Manifold Hypothesis: Hybrid Spectral Parameterizations for Flow Matching
Ségolène Martin, Anne Gagneux, Quentin Bertrand +2
Flow matching and diffusion can be trained to predict different quantities, most commonly the data , the source noise , or the velocity~. Although theoretically equiva…
Training Flow Matching: The Role of Weighting and Parameterization
Anne Gagneux, Ségolène Martin, Rémi Gribonval +1
We study the training objectives of denoising-based generative models, with a particular focus on loss weighting and output parameterization, including noise-, clean image-, and ve…
Reinterpreting EMML as Mirror Descent for Constrained Maximum Likelihood Estimation
Antonin Clerc, Ségolène Martin, Nicolas Papadakis +1
The Expectation--Maximization Maximum Likelihood (EMML) algorithm belongs to the Expectation--Maximization family and is widely used for image reconstruction problems under Poisson…
The Generation Phases of Flow Matching: a Denoising Perspective
Anne Gagneux, Ségolène Martin, Rémi Gribonval +1
Flow matching has achieved remarkable success, yet the factors influencing the quality of its generation process remain poorly understood. In this work, we adopt a denoising perspe…
Learning Brenier Potentials with Convex Generative Adversarial Neural Networks
Claudia Drygala, Hanno Gottschalk, Thomas Kruse +2
Brenier proved that under certain conditions on a source and a target probability measure there exists a strictly convex function such that its gradient is a transport map from the…
Variable Bregman Majorization-Minimization Algorithm and its Application to Dirichlet Maximum Likelihood Estimation
Ségolène Martin, Jean-Christophe Pesquet, Gabriele Steidl +1
We propose a novel Bregman descent algorithm for minimizing a convex function that is expressed as the sum of a differentiable part (defined over an open set) and a possibly nonsmo…
Deep Generative Clustering with VAEs and Expectation-Maximization
Michael Adipoetra, Ségolène Martin
We propose a novel deep clustering method that integrates Variational Autoencoders (VAEs) into the Expectation-Maximization (EM) framework. Our approach models the probability dist…
PnP-Flow: Plug-and-Play Image Restoration with Flow Matching
Ségolène Martin, Anne Gagneux, Paul Hagemann +1
In this paper, we introduce Plug-and-Play (PnP) Flow Matching, an algorithm for solving imaging inverse problems. PnP methods leverage the strength of pre-trained denoisers, often…