Abstract
We introduce a novel framework for WEakly supervised Learning of Deep cOnvolutional neural 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 outperforms state-of-the-art results on six different datasets.
Paper
This paper has been published at IEEE Conference on Computer Vision and Pattern Recognition (CVPR) 2016.
Code with ResNet-101 architecture
Bibtex
@inproceedings{Durand_WELDON_CVPR_2016,
author = {Durand, Thibaut and Thome, Nicolas and Cord, Matthieu},
title = {WELDON: Weakly Supervised Learning of Deep Convolutional Neural Networks},
booktitle = {The IEEE Conference on Computer Vision and Pattern Recognition (CVPR)},
year = {2016}
}
Acknowledgements
This research was supported by a DGA-MRIS scholarship.