Dense Relational Captioning: Triple-Stream Networks for Relationship-Based Captioning
Abstract
Our goal in this work is to train an image captioning model that generates more dense and informative captions. We introduce "relational captioning," a novel image captioning task which aims to generate multiple captions with respect to relational information between objects in an image. Relational captioning is a framework that is advantageous in both diversity and amount of information, leading to image understanding based on relationships. Part-of-speech (POS, i.e. subject-object-predicate categories) tags can be assigned to every English word. We leverage the POS as a prior to guide the correct sequence of words in a caption. To this end, we propose a multi-task triple-stream network (MTTSNet) which consists of three recurrent units for the respective POS and jointly performs POS prediction and captioning. We demonstrate more diverse and richer representations generated by the proposed model against several baselines and competing methods.
Cite
Text
Kim et al. "Dense Relational Captioning: Triple-Stream Networks for Relationship-Based Captioning." Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2019. doi:10.1109/CVPR.2019.00643Markdown
[Kim et al. "Dense Relational Captioning: Triple-Stream Networks for Relationship-Based Captioning." Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2019.](https://mlanthology.org/cvpr/2019/kim2019cvpr-dense/) doi:10.1109/CVPR.2019.00643BibTeX
@inproceedings{kim2019cvpr-dense,
title = {{Dense Relational Captioning: Triple-Stream Networks for Relationship-Based Captioning}},
author = {Kim, Dong-Jin and Choi, Jinsoo and Oh, Tae-Hyun and Kweon, In So},
booktitle = {Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition},
year = {2019},
doi = {10.1109/CVPR.2019.00643},
url = {https://mlanthology.org/cvpr/2019/kim2019cvpr-dense/}
}