Guided Open Vocabulary Image Captioning with Constrained Beam Search

Abstract

Existing image captioning models do not generalize well to out-of-domain images containing novel scenes or objects. This limitation severely hinders the use of these models in real world applications dealing with images in the wild. We address this problem using a flexible approach that enables existing deep captioning architectures to take advantage of… (More)

Topics

12 Figures and Tables

Cite this paper

@inproceedings{Anderson2017GuidedOV, title={Guided Open Vocabulary Image Captioning with Constrained Beam Search}, author={Peter Anderson and Basura Fernando and Mark Johnson and Stephen Gould}, booktitle={EMNLP}, year={2017} }