Frankenstein: Learning Deep Face Representations using Small Data

Abstract : Deep convolutional neural networks have recently proven extremely effective for difficult face recognition problems in uncontrolled settings. To train such networks, very large training sets are needed with millions of labeled images. For some applications, such as near-infrared (NIR) face recognition, such large training datasets are not publicly available and difficult to collect. In this work, we propose a method to generate very large training datasets of synthetic images by compositing real face images in a given dataset. We show that this method enables to learn models from as few as 10,000 training images, which perform on par with models trained from 500,000 images. Using our approach we also obtain state-of-the-art results on the CASIA NIR-VIS2.0 heterogeneous face recognition dataset.
Type de document :
Article dans une revue
IEEE Transactions on Image Processing, Institute of Electrical and Electronics Engineers, 2018, 27 (1), pp.293-303. 〈10.1109/TIP.2017.2756450〉
Liste complète des métadonnées

Littérature citée [67 références]  Voir  Masquer  Télécharger


https://hal.inria.fr/hal-01306168
Contributeur : Thoth Team <>
Soumis le : lundi 25 septembre 2017 - 09:41:55
Dernière modification le : lundi 13 novembre 2017 - 11:57:31

Fichiers

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

Identifiants

Collections

Citation

Guosheng Hu, Xiaojiang Peng, Yongxin Yang, Timothy Hospedales, Jakob Verbeek. Frankenstein: Learning Deep Face Representations using Small Data. IEEE Transactions on Image Processing, Institute of Electrical and Electronics Engineers, 2018, 27 (1), pp.293-303. 〈10.1109/TIP.2017.2756450〉. 〈hal-01306168v2〉

Partager

Métriques

Consultations de
la notice

179

Téléchargements du document

131