This paper proposes a method for learning joint embeddings of images and text using a two-branch neural network with multiple layers of linear projections followed by nonlinearities. The network is trained using a large margin objective that combines cross-view ranking constraints with within-view neighborhood structure preservation constraints inspired by metric learning literature. Extensive experiments show that our approach gains significant improvements in accuracy for image-to-text and text-to-image retrieval. Our method achieves new state-of-the-art results on the Flickr30K and MSCOCO image-sentence datasets and shows promise on the new task of phrase localization on the Flickr30K Entities dataset.
@article{arxiv.1511.06078,
title = {Learning Deep Structure-Preserving Image-Text Embeddings},
author = {Liwei Wang and Yin Li and Svetlana Lazebnik},
journal= {arXiv preprint arXiv:1511.06078},
year = {2016}
}