Explanipedia

Blur2seq: Blind Deblurring and Camera Trajectory Estimation from a Single Camera Motion-blurred Image Open

Guillermo Carbajal, Andrés Almansa, Pablo Musé · 2025

Motion blur caused by camera shake, particularly under large or rotational movements, remains a major challenge in image restoration. We propose a deep learning framework that jointly estimates the latent sharp image and the underlying cam…

LVTINO: LAtent Video consisTency INverse sOlver for High Definition Video Restoration Open

Alessio Spagnoletti, Andrés Almansa, Marcelo Pereyra · 2025

Computational imaging methods increasingly rely on powerful generative diffusion models to tackle challenging image restoration tasks. In particular, state-of-the-art zero-shot image inverse solvers leverage distilled text-to-image latent …

Infusion: Internal Diffusion for Inpainting of Dynamic Textures and Complex Motion Open

Nicolas Cherel, Andrés Almansa, Yann Gousseau, Alasdair Newson · 2025

Video inpainting is the task of filling a region in a video in a visually convincing manner It is very challenging due to the high dimensionality of the data and the temporal consistency required for obtaining convincing results. Recently,…

LATINO-PRO: LAtent consisTency INverse sOlver with PRompt Optimization Open

Alessio Spagnoletti, J.A. Prost, Andrés Almansa, Nicolas Papadakis, Marcelo Pereyra · 2025

Text-to-image latent diffusion models (LDMs) have recently emerged as powerful generative models with great potential for solving inverse problems in imaging. However, leveraging such models in a Plug & Play (PnP), zero-shot manner remains…

Memoria histórica y pensamiento crítico: una experiencia de innovación didáctica sobre el franquismo en un aula de 4º de la ESO Open

Sandra Vazquez, Tania Riveiro Rodríguez, Andrés Almansa · 2024

Este trabajo lleva a cabo un estudio de caso descriptivo en un aula de 4º de la ESO en la que se aborda, desde Geografía e Historia, la cuestión problemática de la memoria histórica a partir de la dictadura franquista. La experiencia didác…

Memoria histórica y pensamiento crítico: una experiencia de innovación didáctica sobre el franquismo en un aula de 4º de la ESO Open

Sandra Vazquez, Tania Riveiro Rodríguez, Andrés Almansa · 2024

Este trabajo lleva a cabo un estudio de caso descriptivo en un aula de 4º de la ESO en la que se aborda, desde Geografía e Historia, la cuestión problemática de la memoria histórica a partir de la dictadura franquista. La experiencia didác…

Hyperspectral Signal Reconstruction from Interferometric Measurements with Enriched Fourier Bases Open

Yassine Mhiri, Rémy Abergel, Andrés Almansa, Lionel Moisan · 2024

International audience

Linear Super-Resolution Through Translational Motion Open

Rémy Abergel, Andrés Almansa, Lionel Moisan, Camille Noûs · 2024

The problem of computing a single super-resolved image from multiple shifted images of the same scene has been addressed in several ways in the literature, with different levels of assumptions on the image to reconstruct, but a frequent co…

Diffusion-based image inpainting with internal learning Open

Nicolas Cherel, Andrés Almansa, Yann Gousseau, Alasdair Newson · 2024

Diffusion models are now the undisputed state-of-the-art for image generation and image restoration. However, they require large amounts of computational power for training and inference. In this paper, we propose lightweight diffusion mod…

Apprentissage d'une fonction de régularisation locale pour la restauration d'images Open

Jean Prost, Antoine Houdard, Andrés Almansa, Nicolas Papadakis · 2024

Dans ce travail, nous élaborons une stratégie afin d’apprendre une fonction de régularisation pour résoudre des problèmes de restauration d’images. Une fonction de régularisation locale, paramétrée par un réseau de neurones convolutif, est…

Impact of image registration errors on the quality of hyperspectral images in imaging static Fourier transform spectrometry Open

Varvara Chiliaeva, Andrés Almansa, Yann Ferrec, Jean‐Michel Gaucel, Olivier Gazzano , et al. · 2024

Imaging static Fourier transform spectrometry (isFTS) is used for pushbroom airborne or spaceborne hyperspectral remote sensing. In isFTS, a static two-wave interferometer imprints linear interference fringes over the image of the scene, s…

Fast Diffusion EM: a diffusion model for blind inverse problems with application to deconvolution Open

Charles Laroche, Andrés Almansa, Eva Coupeté · 2024

Using diffusion models to solve inverse problems is a growing field of research. Current methods assume the degradation to be known and provide impressive results in terms of restoration quality and diversity. In this work, we leverage the…

Efficient Posterior Sampling for Diverse Super-Resolution with Hierarchical VAE Prior Open

Jean Prost, Antoine Houdard, Andrés Almansa, Nicolas Papadakis · 2024

We investigate the problem of producing diverse solutions to an image super-resolution problem.From a probabilistic perspective, this can be done by sampling from the posterior distribution of an inverse problem, which requires the definit…

Infusion: internal diffusion for inpainting of dynamic textures and complex motion Open

Nicolas Cherel, Andrés Almansa, Yann Gousseau, Alasdair Newson · 2023

Video inpainting is the task of filling a region in a video in a visually convincing manner. It is very challenging due to the high dimensionality of the data and the temporal consistency required for obtaining convincing results. Recently…

Patch-based stochastic attention for image editing Open

Nicolas Cherel, Andrés Almansa, Yann Gousseau, Alasdair Newson · 2023

Plug-and-Play Posterior Sampling under Mismatched Measurement and Prior Models Open

Marien Renaud, Jiaming Liu, Valentin De Bortoli, Andrés Almansa, Ulugbek S. Kamilov · 2023

Posterior sampling has been shown to be a powerful Bayesian approach for solving imaging inverse problems. The recent plug-and-play unadjusted Langevin algorithm (PnP-ULA) has emerged as a promising method for Monte Carlo sampling and mini…

Modèle de diffusion frugal pour l'inpainting d'images Open

Nicolas Cherel, Andrés Almansa, Yann Gousseau, Alasdair Newson · 2023

National audience

Imaging static Fourier transform spectrometry: impact of trajectory perturbations on the hyperspectral images Open

Varvara Chiliaeva, Olivier Gazzano, Yann Ferrec, Hervé Sauer, Andrés Almansa , et al. · 2023

Imaging static Fourier transform spectrometry (isFTS) may be used for pushbroom air-or spaceborne hyperspectral remote sensing.In isFTS, the spectral information is multiplexed over several instantaneous images, and numerical reconstructio…

«epidemia» del sport. Educación, socialización y expansión del deporte en Galicia y Cataluña (1870-1914) Open

Roque Sanfiz Arias, Andrés Almansa · 2023

Desde la perspectiva comparada analizamos la introducción y el arraigo social del fenómeno deportivo en Galicia y Cataluña. El proceso comenzó en los gimnasios, iniciando un modelo asociativo que paliaba la ausencia de cultura física en el…

Inverse problem regularization with hierarchical variational autoencoders Open

J.A. Prost, Antoine Houdard, Andrés Almansa, Nicolas Papadakis · 2023

In this paper, we propose to regularize ill-posed inverse problems using a deep hierarchical variational autoencoder (HVAE) as an image prior. The proposed method synthesizes the advantages of i) denoiser-based Plug \& Play approaches and …

Deep Model-Based Super-Resolution with Non-uniform Blur Open

Charles Laroche, Andrés Almansa, Matias Tassano · 2023

We propose a state-of-the-art method for super-resolution with non-uniform\nblur. Single-image super-resolution methods seek to restore a high-resolution\nimage from blurred, subsampled, and noisy measurements. Despite their\nimpressive pe…

On Maximum a Posteriori Estimation with Plug & Play Priors and Stochastic Gradient Descent Open

Rémi Laumont, Valentin De Bortoli, Andrés Almansa, Julie Delon, Alain Durmus , et al. · 2023

Provably Convergent Plug & Play Linearized ADMM, applied to Deblurring Spatially Varying Kernels Open

Charles Laroche, Andrés Almansa, Eva Coupeté, Matias Tassano · 2022

Plug & Play methods combine proximal algorithms with denoiser priors to solve inverse problems. These methods rely on the computability of the proximal operator of the data fidelity term. In this paper, we propose a Plug & Play framework b…

Video Restoration with a Deep Plug-and-Play Prior Open

Antoine Monod, Julie Delon, Matias Tassano, Andrés Almansa · 2022

This paper presents a novel method for restoring digital videos via a Deep Plug-and-Play (PnP) approach. Under a Bayesian formalism, the method consists in using a deep convolutional denoising network in place of the proximal operator of t…

Solving Inverse Problems by Joint Posterior Maximization with Autoencoding Prior Open

Mario González, Andrés Almansa, Pauline Tan · 2022

International audience

Bayesian Imaging Using Plug & Play Priors: When Langevin Meets Tweedie Open

Rémi Laumont, Valentin De Bortoli, Andrés Almansa, Julie Delon, Alain Durmus , et al. · 2022

International audience

Diverse super-resolution with pretrained deep hiererarchical VAEs Open

J.A. Prost, Antoine Houdard, Nicolas Papadakis, Andrés Almansa · 2022

We investigate the problem of producing diverse solutions to an image super-resolution problem. From a probabilistic perspective, this can be done by sampling from the posterior distribution of an inverse problem, which requires the defini…

On Maximum-a-Posteriori estimation with Plug & Play priors and stochastic gradient descent Open

Rémi Laumont, Valentin De Bortoli, Andrés Almansa, Julie Delon, Alain Durmus , et al. · 2022

Bayesian methods to solve imaging inverse problems usually combine an explicit data likelihood function with a prior distribution that explicitly models expected properties of the solution. Many kinds of priors have been explored in the li…

Diverse Super-Resolution with Pretrained Hierarchical Variational Autoencoders Open

Jean Prost, Antoine Houdard, Andrés Almansa, Nicolas Papadakis · 2022

Bayesian imaging using Plug & Play priors: when Langevin meets Tweedie Open

Rémi Laumont, Valentin De Bortoli, Andrés Almansa, Julie Delon, Alain Durmus , et al. · 2021

Since the seminal work of Venkatakrishnan et al. in 2013, Plug & Play (PnP) methods have become ubiquitous in Bayesian imaging. These methods derive Minimum Mean Square Error (MMSE) or Maximum A Posteriori (MAP) estimators for inverse prob…

Andrés Almansa YOU? Author Swipe