Image Captioning with Deep Bidirectional LSTMs
This work addresses image captioning for computer vision applications, offering incremental improvements by enhancing LSTM depth and using data augmentation without new mechanisms like attention.
The authors tackled image captioning by proposing deep bidirectional LSTM models that learn visual-language interactions, achieving competitive performance on caption generation and significantly outperforming recent methods on image-sentence retrieval tasks across benchmark datasets like Flickr8K, Flickr30K, and MSCOCO.
This work presents an end-to-end trainable deep bidirectional LSTM (Long-Short Term Memory) model for image captioning. Our model builds on a deep convolutional neural network (CNN) and two separate LSTM networks. It is capable of learning long term visual-language interactions by making use of history and future context information at high level semantic space. Two novel deep bidirectional variant models, in which we increase the depth of nonlinearity transition in different way, are proposed to learn hierarchical visual-language embeddings. Data augmentation techniques such as multi-crop, multi-scale and vertical mirror are proposed to prevent overfitting in training deep models. We visualize the evolution of bidirectional LSTM internal states over time and qualitatively analyze how our models "translate" image to sentence. Our proposed models are evaluated on caption generation and image-sentence retrieval tasks with three benchmark datasets: Flickr8K, Flickr30K and MSCOCO datasets. We demonstrate that bidirectional LSTM models achieve highly competitive performance to the state-of-the-art results on caption generation even without integrating additional mechanism (e.g. object detection, attention model etc.) and significantly outperform recent methods on retrieval task.