Unsupervised Learning of Artistic Styles with Archetypal Style Analysis

Abstract : In this paper, we introduce an unsupervised learning approach to automatically discover , summarize, and manipulate artistic styles from large collections of paintings. Our method is based on archetypal analysis, which is an unsupervised learning technique akin to sparse coding with a geometric interpretation. When applied to deep image representations from a collection of artworks, it learns a dictionary of archetypal styles, which can be easily visualized. After training the model, the style of a new image, which is characterized by local statistics of deep visual features, is approximated by a sparse convex combination of archetypes. This enables us to interpret which archetypal styles are present in the input image, and in which proportion. Finally, our approach allows us to manipulate the coefficients of the latent archetypal decomposition, and achieve various special effects such as style enhancement, transfer, and interpolation between multiple archetypes.
Type de document :
Pré-publication, Document de travail
Accepted at NIPS 2018, Montréal, Canada. 2018
Liste complète des métadonnées

https://hal.inria.fr/hal-01802131
Contributeur : Daan Wynen <>
Soumis le : mercredi 3 octobre 2018 - 11:39:15
Dernière modification le : samedi 6 octobre 2018 - 01:08:32

Fichier

archetypal_style.pdf
Fichiers produits par l'(les) auteur(s)

Identifiants

  • HAL Id : hal-01802131, version 2

Collections

INRIA | LJK | UGA

Citation

Daan Wynen, Cordelia Schmid, Julien Mairal. Unsupervised Learning of Artistic Styles with Archetypal Style Analysis. Accepted at NIPS 2018, Montréal, Canada. 2018. 〈hal-01802131v2〉

Partager

Métriques

Consultations de la notice

70

Téléchargements de fichiers

43