Regularization via Deep Generative Models: an Analysis Point of View - Archive ouverte HAL Accéder directement au contenu
Communication Dans Un Congrès Année : 2021

Regularization via Deep Generative Models: an Analysis Point of View

Résumé

This paper proposes a new way of regularizing an inverse problem in imaging (e.g., deblurring or inpainting) by means of a deep generative neural network. Compared to end-to-end models, such approaches seem particularly interesting since the same network can be used for many different problems and experimental conditions, as soon as the generative model is suited to the data. Previous works proposed to use a synthesis framework, where the estimation is performed on the latent vector, the solution being obtained afterwards via the decoder. Instead, we propose an analysis formulation where we directly optimize the image itself and penalize the latent vector. We illustrate the interest of such a formulation by running experiments of inpainting, deblurring and super-resolution. In many cases our technique achieves a clear improvement of the performance and seems to be more robust, in particular with respect to initialization.
Fichier principal
Vignette du fichier
Oberlin_28453.pdf (737.77 Ko) Télécharger le fichier
Origine : Fichiers produits par l'(les) auteur(s)

Dates et versions

hal-03390697 , version 1 (21-10-2021)

Identifiants

Citer

Thomas Oberlin, Mathieu Verm. Regularization via Deep Generative Models: an Analysis Point of View. 28th IEEE International Conference on Image Processing (ICIP 2021), IEEE Signal Processing Society, Sep 2021, Anchorage, United States. pp.404-408, ⟨10.1109/ICIP42928.2021.9506138⟩. ⟨hal-03390697⟩
48 Consultations
73 Téléchargements

Altmetric

Partager

Gmail Facebook X LinkedIn More