Multimodal Compact Bilinear Pooling for Multimodal Neural Machine Translation
This work addresses multimodal translation for tasks like image caption translation, but it is incremental as it applies an existing pooling method from visual question answering to a new domain.
The paper tackled the problem of combining visual and textual features in multimodal neural machine translation by evaluating Multimodal Compact Bilinear pooling, which improved translation performance compared to basic methods like element-wise product or concatenation.
In state-of-the-art Neural Machine Translation, an attention mechanism is used during decoding to enhance the translation. At every step, the decoder uses this mechanism to focus on different parts of the source sentence to gather the most useful information before outputting its target word. Recently, the effectiveness of the attention mechanism has also been explored for multimodal tasks, where it becomes possible to focus both on sentence parts and image regions. Approaches to pool two modalities usually include element-wise product, sum or concatenation. In this paper, we evaluate the more advanced Multimodal Compact Bilinear pooling method, which takes the outer product of two vectors to combine the attention features for the two modalities. This has been previously investigated for visual question answering. We try out this approach for multimodal image caption translation and show improvements compared to basic combination methods.