Cross-Domain Fashion Image Retrieval
Abstract
Cross domain image retrieval is a challenging task that implies matching images from one domain to their pairs from another domain. In this paper we focus on fashion image retrieval, which involves matching an image of a fashion item taken by users, to the images of the same item taken in controlled condition, usually by professional photographer. When facing this problem, we have different products in train and test time, and we use triplet loss to train the network. We stress the importance of proper training of simple architecture, as well as adapting general models to the specific task.
Cite
Text
Gajic and Baldrich. "Cross-Domain Fashion Image Retrieval." IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops, 2018. doi:10.1109/CVPRW.2018.00243Markdown
[Gajic and Baldrich. "Cross-Domain Fashion Image Retrieval." IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops, 2018.](https://mlanthology.org/cvprw/2018/gajic2018cvprw-crossdomain/) doi:10.1109/CVPRW.2018.00243BibTeX
@inproceedings{gajic2018cvprw-crossdomain,
title = {{Cross-Domain Fashion Image Retrieval}},
author = {Gajic, Bojana and Baldrich, Ramón},
booktitle = {IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops},
year = {2018},
pages = {1869-1871},
doi = {10.1109/CVPRW.2018.00243},
url = {https://mlanthology.org/cvprw/2018/gajic2018cvprw-crossdomain/}
}