Hard Negative Mining for Metric Learning Based Zero-Shot Classification
Abstract
Zero-Shot learning has been shown to be an efficient strategy for domain adaptation. In this context, this paper builds on the recent work of Bucher et al. [1], which proposed an approach to solve Zero-Shot classification problems (ZSC) by introducing a novel metric learning based objective function. This objective function allows to learn an optimal embedding of the attributes jointly with a measure of similarity between images and attributes. This paper extends their approach by proposing several schemes to control the generation of the negative pairs, resulting in a significant improvement of the performance and giving above state-of-the-art results on three challenging ZSC datasets.
Cite
Text
Bucher et al. "Hard Negative Mining for Metric Learning Based Zero-Shot Classification." European Conference on Computer Vision Workshops, 2016. doi:10.1007/978-3-319-49409-8_45Markdown
[Bucher et al. "Hard Negative Mining for Metric Learning Based Zero-Shot Classification." European Conference on Computer Vision Workshops, 2016.](https://mlanthology.org/eccvw/2016/bucher2016eccvw-hard/) doi:10.1007/978-3-319-49409-8_45BibTeX
@inproceedings{bucher2016eccvw-hard,
title = {{Hard Negative Mining for Metric Learning Based Zero-Shot Classification}},
author = {Bucher, Maxime and Herbin, Stéphane and Jurie, Frédéric},
booktitle = {European Conference on Computer Vision Workshops},
year = {2016},
pages = {524-531},
doi = {10.1007/978-3-319-49409-8_45},
url = {https://mlanthology.org/eccvw/2016/bucher2016eccvw-hard/}
}