In this paper, we propose a novel conditional-generative-adversarial-nets-based image captioning framework as an extension of traditional reinforcement-learning (RL)-based encoder-decoder architecture. To deal with the inconsistent evaluation problem among different objective language metrics, we are motivated to design some "discriminator" networks to automatically and progressively determine whether generated caption is human described or machine generated. Two kinds of discriminator architectures (CNN and RNN-based structures) are introduced since each has its own advantages. The proposed algorithm is generic so that it can enhance any existing RL-based image captioning framework and we show that the conventional RL training method is just a special case of our approach. Empirically, we show consistent improvements over all language evaluation metrics for different state-of-the-art image captioning models. In addition, the well-trained discriminators can also be viewed as objective image captioning evaluators
@article{arxiv.1805.07112,
title = {Improving Image Captioning with Conditional Generative Adversarial Nets},
author = {Chen Chen and Shuai Mu and Wanpeng Xiao and Zexiong Ye and Liesi Wu and Qi Ju},
journal= {arXiv preprint arXiv:1805.07112},
year = {2020}
}
Comments
12 pages; 33 figures; 36 refenences; Accepted by AAAI2019