pith. sign in

arxiv: 1910.14208 · v2 · pith:2ATDKVU4new · submitted 2019-10-31 · 💻 cs.CV · cs.CL

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

classification 💻 cs.CV cs.CL
keywords hiddenimagecaptionscaptioncaptioningconditioneddecodergenerated
0
0 comments X
read the original 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.

This paper has not been read by Pith yet.

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.