Joint CNN and Variational Model for Fully-automatic Image Colorization - Archive ouverte HAL Accéder directement au contenu
Communication Dans Un Congrès Année : 2019

Joint CNN and Variational Model for Fully-automatic Image Colorization

Résumé

This paper aims to couple the powerful prediction of the convolutional neural network (CNN) to the accuracy at pixel scale of the variational methods. In this work, the limitations of the CNN-based image colorization approaches are described. We then focus on a CNN which is able to compute a statistical distribution of the colors for each pixel of the image based on a learning over a large color image database. After describing its limitation, the variational method of [17] is briefly recalled. This method is able to select a color candidate among a given set while performing a regularization of the result. By combining this approach with a CNN, we designed a fully automatic image coloriza-tion framework with an improved accuracy in comparison with CNN alone. Some numerical experiments demonstrate the increased accuracy performed by our method.
Fichier principal
Vignette du fichier
pierre2019ssvm.pdf (6.97 Mo) Télécharger le fichier
Origine : Fichiers produits par l'(les) auteur(s)
Loading...

Dates et versions

hal-02059820 , version 1 (06-03-2019)
hal-02059820 , version 2 (14-03-2019)

Identifiants

  • HAL Id : hal-02059820 , version 2

Citer

Thomas Mouzon, Fabien Pierre, Marie-Odile Berger. Joint CNN and Variational Model for Fully-automatic Image Colorization. SSVM 2019 - Seventh International Conference on Scale Space and Variational Methods in Computer Vision, Jun 2019, Hofgeismar, Germany. pp.535-546. ⟨hal-02059820v2⟩
490 Consultations
369 Téléchargements

Partager

Gmail Facebook X LinkedIn More