no code implementations • 2 Nov 2023 • Nicolas Cherel, Andrés Almansa, Yann Gousseau, Alasdair Newson
We show that in the case of video inpainting, thanks to the highly auto-similar nature of videos, the training of a diffusion model can be restricted to the video to inpaint and still produce very satisfying results.
1 code implementation • 5 Oct 2023 • Marien Renaud, Jiaming Liu, Valentin De Bortoli, Andrés Almansa, Ulugbek S. Kamilov
Posterior sampling has been shown to be a powerful Bayesian approach for solving imaging inverse problems.
1 code implementation • 1 Sep 2023 • Charles Laroche, Andrés Almansa, Eva Coupete
Our method alternates between approximating the expected log-likelihood of the inverse problem using samples drawn from a diffusion model and a maximization step to estimate unknown model parameters.
1 code implementation • ICCV 2023 • Jean Prost, Antoine Houdard, Andrés Almansa, Nicolas Papadakis
Our experiments show that the proposed PnP-HVAE method is competitive with both SOTA denoiser-based PnP approaches, and other SOTA restoration methods based on generative models.
1 code implementation • 19 Oct 2022 • Charles Laroche, Andrés Almansa, Eva Coupeté, Matias Tassano
Plug & Play methods combine proximal algorithms with denoiser priors to solve inverse problems.
no code implementations • 6 Sep 2022 • Antoine Monod, Julie Delon, Matias Tassano, Andrés Almansa
Under a Bayesian formalism, the method consists in using a deep convolutional denoising network in place of the proximal operator of the prior in an alternating optimization scheme.
no code implementations • 20 May 2022 • Jean Prost, Antoine Houdard, Andrés Almansa, Nicolas Papadakis
We investigate the problem of producing diverse solutions to an image super-resolution problem.
2 code implementations • 21 Apr 2022 • Charles Laroche, Andrés Almansa, Matias Tassano
Instead, in this paper, we address the more realistic and computationally challenging case of spatially-varying blur.
1 code implementation • 7 Feb 2022 • Nicolas Cherel, Andrés Almansa, Yann Gousseau, Alasdair Newson
We refer to our proposed layer as a "Patch-based Stochastic Attention Layer" (PSAL).
no code implementations • 16 Jan 2022 • Rémi Laumont, Valentin De Bortoli, Andrés Almansa, Julie Delon, Alain Durmus, Marcelo Pereyra
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.
no code implementations • 8 Mar 2021 • Rémi Laumont, Valentin De Bortoli, Andrés Almansa, Julie Delon, Alain Durmus, Marcelo Pereyra
The proposed algorithms are demonstrated on several canonical problems such as image deblurring, inpainting, and denoising, where they are used for point estimation as well as for uncertainty visualisation and quantification.
1 code implementation • 2 Mar 2021 • Mario González, Andrés Almansa, Pauline Tan
Whereas previous MAP-based approaches to this problem lead to highly non-convex optimization algorithms, our approach computes the joint (space-latent) MAP that naturally leads to alternate optimization algorithms and to the use of a stochastic encoder to accelerate computations.
no code implementations • 11 Feb 2021 • Jean Prost, Antoine Houdard, Andrés Almansa, Nicolas Papadakis
In this work, we propose a framework to learn a local regularization model for solving general image restoration problems.
1 code implementation • 18 Nov 2019 • Marcela Carvalho, Bertrand Le Saux, Pauline Trouvé-Peloux, Frédéric Champagnat, Andrés Almansa
Aerial or satellite imagery is a great source for land surface analysis, which might yield land use maps or elevation models.
1 code implementation • 14 Nov 2019 • Mario González, Andrés Almansa, Mauricio Delbracio, Pablo Musé, Pauline Tan
In this paper we address the problem of solving ill-posed inverse problems in imaging where the prior is a neural generative model.
no code implementations • 15 Apr 2019 • Alasdair Newson, Andrés Almansa, Yann Gousseau, Saïd Ladjal
This results in a wide range of practical problems, such as difficulties in training, the tendency to sample images with little or no variability, and generalisation problems.
1 code implementation • 5 Sep 2018 • Marcela Carvalho, Bertrand Le Saux, Pauline Trouvé-Peloux, Andrés Almansa, Frédéric Champagnat
We also investigate the influence of blur on depth prediction observing model uncertainty with a Bayesian neural network approach.
no code implementations • 10 Jun 2017 • Cecilia Aguerrebere, Andrés Almansa, Julie Delon, Yann Gousseau, Pablo Musé
In this work, we propose the use of a hyperprior to model image patches, in order to stabilize the estimation procedure.
no code implementations • 18 Mar 2015 • Alasdair Newson, Andrés Almansa, Matthieu Fradet, Yann Gousseau, Patrick Pérez
Our algorithm is able to deal with a variety of challenging situations which naturally arise in video inpainting, such as the correct reconstruction of dynamic textures, multiple moving objects and moving background.