Swapping Semantic Contents for Mixing Images - CNRS - Centre national de la recherche scientifique Accéder directement au contenu
Communication Dans Un Congrès Année : 2022

Swapping Semantic Contents for Mixing Images

Résumé

Deep architecture have proven capable of solving many tasks provided a sufficient amount of labeled data. In fact, the amount of available labeled data has become the principal bottleneck in low label settings such as Semi-Supervised Learning. Mixing Data Augmentations do not typically yield new labeled samples, as indiscriminately mixing contents creates between-class samples. In this work, we introduce the SciMix framework that can learn to replace the global semantic content from one sample. By teaching a StyleGan generator to embed a semantic style code into image backgrounds, we obtain new mixing scheme for data augmentation. We then demonstrate that SciMix yields novel mixed samples that inherit many characteristics from their non-semantic parents. Afterwards, we verify those samples can be used to improve the performance semi-supervised frameworks like Mean Teacher or Fixmatch, and even fully supervised learning on a small labeled dataset.
Fichier principal
Vignette du fichier
2205.10158.pdf (8.24 Mo) Télécharger le fichier
Origine : Fichiers produits par l'(les) auteur(s)

Dates et versions

hal-03951744 , version 1 (23-01-2023)

Identifiants

Citer

Rémy Sun, Clément Masson, Gilles Hénaff, Nicolas Thome, Matthieu Cord. Swapping Semantic Contents for Mixing Images. 2022 26th International Conference on Pattern Recognition (ICPR), Aug 2022, Montreal, Canada. pp.1280-1286, ⟨10.1109/ICPR56361.2022.9956602⟩. ⟨hal-03951744⟩
13 Consultations
14 Téléchargements

Altmetric

Partager

Gmail Facebook X LinkedIn More