Deep Captioning with Multimodal Recurrent Neural Networks (m-RNN)

TLDR

The paper proposes a multimodal Recurrent Neural Network (m‑RNN) for generating novel image captions. The m‑RNN jointly models word probabilities conditioned on prior words and an image using a deep RNN for language and a deep CNN for vision, with a multimodal layer, and is evaluated on IAPR TC‑12, Flickr 8K/30K, and MS COCO. The m‑RNN outperforms state‑of‑the‑art captioning methods and yields significant gains on image‑sentence retrieval. Project page: www.stat.ucla.edu/~junhua.mao/m‑RNN.html.

Abstract

In this paper, we present a multimodal Recurrent Neural Network (m-RNN) model for generating novel image captions. It directly models the probability distribution of generating a word given previous words and an image. Image captions are generated by sampling from this distribution. The model consists of two sub-networks: a deep recurrent neural network for sentences and a deep convolutional network for images. These two sub-networks interact with each other in a multimodal layer to form the whole m-RNN model. The effectiveness of our model is validated on four benchmark datasets (IAPR TC-12, Flickr 8K, Flickr 30K and MS COCO). Our model outperforms the state-of-the-art methods. In addition, we apply the m-RNN model to retrieval tasks for retrieving images or sentences, and achieves significant performance improvement over the state-of-the-art methods which directly optimize the ranking objective function for retrieval. The project page of this work is: www.stat.ucla.edu/~junhua.mao/m-RNN.html .