WILDCAT: Weakly Supervised Learning of Deep ConvNets for Image Classification, Pointwise Localization and Segmentation

Abstract : This paper introduces WILDCAT, a deep learning method which jointly aims at aligning image regions for gaining spatial invariance and learning strongly localized features. Our model is trained using only global image labels and is devoted to three main visual recognition tasks: image classification, weakly supervised pointwise object lo-calization and semantic segmentation. WILDCAT extends state-of-the-art Convolutional Neural Networks at three major levels: the use of Fully Convolutional Networks for maintaining spatial resolution, the explicit design in the network of local features related to different class modalities, and a new way to pool these features to provide a global image prediction required for weakly supervised training. Extensive experiments show that our model significantly out-performs the state-of-the-art methods.
Liste complète des métadonnées

Cited literature [65 references]  Display  Hide  Download

https://hal.archives-ouvertes.fr/hal-01515640
Contributor : Thibaut Durand <>
Submitted on : Tuesday, August 29, 2017 - 1:58:03 PM
Last modification on : Thursday, March 21, 2019 - 2:44:14 PM
Document(s) archivé(s) le : Wednesday, January 10, 2018 - 11:01:57 AM

File

0200.pdf
Files produced by the author(s)

Identifiers

  • HAL Id : hal-01515640, version 1

Citation

Thibaut Durand, Taylor Mordan, Nicolas Thome, Matthieu Cord. WILDCAT: Weakly Supervised Learning of Deep ConvNets for Image Classification, Pointwise Localization and Segmentation. IEEE Conference on Computer Vision and Pattern Recognition (CVPR 2017), IEEE, Jul 2017, Honolulu, HI, United States. ⟨hal-01515640⟩

Share

Metrics

Record views

843

Files downloads

366