Multimodal Unified Attention Networks for Vision-and-Language Interactions
This addresses the need for better multimodal understanding in AI applications like visual question answering, though it is incremental by extending existing co-attention models.
The paper tackled the problem of learning effective attention for multimodal data in vision-and-language tasks by proposing a unified attention model that captures both intra- and inter-modal interactions, achieving top-level performance on VQA and visual grounding datasets.
Learning an effective attention mechanism for multimodal data is important in many vision-and-language tasks that require a synergic understanding of both the visual and textual contents. Existing state-of-the-art approaches use co-attention models to associate each visual object (e.g., image region) with each textual object (e.g., query word). Despite the success of these co-attention models, they only model inter-modal interactions while neglecting intra-modal interactions. Here we propose a general `unified attention' model that simultaneously captures the intra- and inter-modal interactions of multimodal features and outputs their corresponding attended representations. By stacking such unified attention blocks in depth, we obtain the deep Multimodal Unified Attention Network (MUAN), which can seamlessly be applied to the visual question answering (VQA) and visual grounding tasks. We evaluate our MUAN models on two VQA datasets and three visual grounding datasets, and the results show that MUAN achieves top-level performance on both tasks without bells and whistles.