Sequentially Generated Instance-Dependent Image Representations for Classification

Abstract : In this paper, we investigate a new framework for image classification that adaptively generates spatial representations. Our strategy is based on a sequential process that learns to explore the different regions of any image in order to infer its category. In particular, the choice of regions is specific to each image, directed by the actual content of previously selected regions.The capacity of the system to handle incomplete image information as well as its adaptive region selection allow the system to perform well in budgeted classification tasks by exploiting a dynamicly generated representation of each image. We demonstrate the system's abilities in a series of image-based exploration and classification tasks that highlight its learned exploration and inference abilities.
Type de document :
Communication dans un congrès
International Conference on Learning Representations, ICLR 2014, Apr 2014, Banff, Canada
Liste complète des métadonnées

https://hal.archives-ouvertes.fr/hal-01215181
Contributeur : Lip6 Publications <>
Soumis le : mardi 13 octobre 2015 - 16:31:59
Dernière modification le : jeudi 22 novembre 2018 - 15:05:10

Identifiants

  • HAL Id : hal-01215181, version 1

Collections

Citation

Gabriel Dulac-Arnold, Ludovic Denoyer, Nicolas Thome, Matthieu Cord, Patrick Gallinari. Sequentially Generated Instance-Dependent Image Representations for Classification. International Conference on Learning Representations, ICLR 2014, Apr 2014, Banff, Canada. 〈hal-01215181〉

Partager

Métriques

Consultations de la notice

58