We present a new technique for learning visual-semantic embeddings for cross-modal retrieval. Inspired by hard negative mining, the use of hard negatives in structured prediction, and ranking loss functions, we introduce a simple change to common loss functions used for multi-modal embeddings. That, combined with fine-tuning and use of augmented data, yields significant gains in retrieval performance. We showcase our approach, VSE++, on MS-COCO and Flickr30K datasets, using ablation studies and comparisons with existing methods. On MS-COCO our approach outperforms state-of-the-art methods by 8.8% in caption retrieval and 11.3% in image retrieval (at R@1).
@article{arxiv.1707.05612,
title = {VSE++: Improving Visual-Semantic Embeddings with Hard Negatives},
author = {Fartash Faghri and David J. Fleet and Jamie Ryan Kiros and Sanja Fidler},
journal= {arXiv preprint arXiv:1707.05612},
year = {2018}
}
Comments
Accepted as spotlight presentation at British Machine Vision Conference (BMVC) 2018. Code: https://github.com/fartashf/vsepp