Context-Gan: Controllable Context Image Generation Using Gans - Aix-Marseille Université
Communication Dans Un Congrès Année : 2023

Context-Gan: Controllable Context Image Generation Using Gans

Résumé

We propose an enhancement to label-to-image GANs. Based on a Pix2Pix architecture, ConText-GAN allows to generate images in a controlled way. Given a feature map as input, ConText-GAN can generate images with a specified layout and label content. As an application, ConText-GAN is used to perform a more realistic than usual data augmentation from an MRI dataset. We show the validity of the generated images with respect to the input feature maps. The relevance of the approach is demonstrated by the improvement of the segmentation result following a data augmentation performed with ConText-GAN compared to classical methods. A practical application is presented in the challenging context of U-Net segmentation of MRI of fat infiltrated muscles.
Fichier principal
Vignette du fichier
ISBI23_paper_653.pdf (1.11 Mo) Télécharger le fichier
Origine Fichiers produits par l'(les) auteur(s)

Dates et versions

hal-04350685 , version 1 (13-03-2024)

Identifiants

Citer

Marc-Adrien Hostin, Vladimir Sivtsov, Shahram Attarian, David Bendahan, Marc-Emmanuel Bellemare. Context-Gan: Controllable Context Image Generation Using Gans. 2023 IEEE 20th International Symposium on Biomedical Imaging (ISBI), Apr 2023, Cartagena, Colombia. ⟨10.1109/ISBI53787.2023.10230602⟩. ⟨hal-04350685⟩
37 Consultations
58 Téléchargements

Altmetric

Partager

More