Deep Feature Factorization for Concept Discovery

We propose Deep Feature Factorization (DFF), a method capable of localizing similar semantic concepts within an image or a set of images. We use DFF to gain insight into a deep convolutional neural network's learned features, where we detect hierarchical cluster structures in feature space. This is visualized as heat maps, which highlight semantically matching regions across a set of images, revealing what the network 'perceives' as similar. DFF can also be used to perform co-segmentation and co-localization, and we report state-of-the-art results on these tasks.


Publié dans:
Proceedings of the 15th European Conference on Computer Vision, 14, 352-368
Présenté à:
15th European Conference on Computer Vision (ECCV), Munich, Germany, September 8-11, 2018
Année
2018
Publisher:
SpringerLink
ISBN:
978-3-030-01267-0
Mots-clefs:
Laboratoires:




 Notice créée le 2019-01-10, modifiée le 2019-03-16

Fichiers:
Télécharger le document
PDF

Évaluer ce document:

Rate this document:
1
2
3
 
(Pas encore évalué)