WELDON: Weakly Supervised Learning of Deep Convolutional Neural Networks - Archive ouverte HAL Access content directly
Conference Papers Year :

WELDON: Weakly Supervised Learning of Deep Convolutional Neural Networks

Abstract

In this paper, we introduce a novel framework for WEakly supervised Learning of Deep cOnvolutional neu-ral Networks (WELDON). Our method is dedicated to automatically selecting relevant image regions from weak annotations , e.g. global image labels, and encompasses the following contributions. Firstly, WELDON leverages recent improvements on the Multiple Instance Learning paradigm, i.e. negative evidence scoring and top instance selection. Secondly, the deep CNN is trained to optimize Average Precision , and fine-tuned on the target dataset with efficient computations due to convolutional feature sharing. A thorough experimental validation shows that WELDON outper-forms state-of-the-art results on six different datasets.
Fichier principal
Vignette du fichier
weldon_cvpr2016.pdf (1.58 Mo) Télécharger le fichier
Origin : Files produced by the author(s)
Loading...

Dates and versions

hal-01343785 , version 1 (10-07-2016)

Identifiers

  • HAL Id : hal-01343785 , version 1

Cite

Thibaut Durand, Nicolas Thome, Matthieu Cord. WELDON: Weakly Supervised Learning of Deep Convolutional Neural Networks. 29th IEEE Conference on Computer Vision and Pattern Recognition (CVPR 2016), Jun 2016, Las Vegas, NV, United States. ⟨hal-01343785⟩
443 View
307 Download

Share

Gmail Facebook Twitter LinkedIn More