Back to Search Start Over

Hidden State Guidance: Improving Image Captioning using An Image Conditioned Autoencoder

Authors :
Wu, Jialin
Mooney, Raymond J.
Publication Year :
2019
Publisher :
arXiv, 2019.

Abstract

Most RNN-based image captioning models receive supervision on the output words to mimic human captions. Therefore, the hidden states can only receive noisy gradient signals via layers of back-propagation through time, leading to less accurate generated captions. Consequently, we propose a novel framework, Hidden State Guidance (HSG), that matches the hidden states in the caption decoder to those in a teacher decoder trained on an easier task of autoencoding the captions conditioned on the image. During training with the REINFORCE algorithm, the conventional rewards are sentence-based evaluation metrics equally distributed to each generated word, no matter their relevance. HSG provides a word-level reward that helps the model learn better hidden representations. Experimental results demonstrate that HSG clearly outperforms various state-of-the-art caption decoders using either raw images or detected objects as inputs.

Details

Database :
OpenAIRE
Accession number :
edsair.doi.dedup.....7743723ee0c90c2700ffb32343cb691a
Full Text :
https://doi.org/10.48550/arxiv.1910.14208