English

Improving Image Captioning by Leveraging Knowledge Graphs

Computer Vision and Pattern Recognition 2019-01-28 v1

Abstract

We explore the use of a knowledge graphs, that capture general or commonsense knowledge, to augment the information extracted from images by the state-of-the-art methods for image captioning. The results of our experiments, on several benchmark data sets such as MS COCO, as measured by CIDEr-D, a performance metric for image captioning, show that the variants of the state-of-the-art methods for image captioning that make use of the information extracted from knowledge graphs can substantially outperform those that rely solely on the information extracted from images.

Keywords

Cite

@article{arxiv.1901.08942,
  title  = {Improving Image Captioning by Leveraging Knowledge Graphs},
  author = {Yimin Zhou and Yiwei Sun and Vasant Honavar},
  journal= {arXiv preprint arXiv:1901.08942},
  year   = {2019}
}

Comments

Accepted by WACV'19

R2 v1 2026-06-23T07:22:22.633Z