Convolutional Transform learning

Abstract : This work proposes a new representation learning technique called convolutional transform learning. In standard transform learning, a dense basis is learned that analyses the image to generate the representation from the image. Here, we learn a set of independent convolutional filters that operate on the images to produce representations (one corresponding to each filter). The major advantage of our proposed approach is that it is completely unsupervised; unlike CNNs where labeled images are required for training. Moreover, it relies on a well-sounded minimization technique with established convergence guarantees. We have compared the proposed method with dictionary learning and transform learning on standard image classification datasets. Results show that our method improves over the rest by a considerable margin.
Type de document :
Communication dans un congrès
25th International Conference on Neural Information Processing (ICONIP 2018), Dec 2018, Siem Reap, Cambodia. 11303, pp.162-174, 2018, Lecture Notes in Computer Science
Liste complète des métadonnées

Littérature citée [20 références]  Voir  Masquer  Télécharger

https://hal.archives-ouvertes.fr/hal-01862201
Contributeur : Emilie Chouzenoux <>
Soumis le : mercredi 28 novembre 2018 - 10:29:12
Dernière modification le : samedi 22 décembre 2018 - 07:23:03

Fichier

typeinst1.pdf
Fichiers produits par l'(les) auteur(s)

Identifiants

  • HAL Id : hal-01862201, version 1

Citation

Jyoti Maggu, Emilie Chouzenoux, Giovanni Chierchia, Angshul Majumdar. Convolutional Transform learning. 25th International Conference on Neural Information Processing (ICONIP 2018), Dec 2018, Siem Reap, Cambodia. 11303, pp.162-174, 2018, Lecture Notes in Computer Science. 〈hal-01862201〉

Partager

Métriques

Consultations de la notice

22

Téléchargements de fichiers

59