Cross-modal Relational Reasoning Network for Visual Question Answering

Visual Question Answering (VQA) is a challenging task that requires a cross-modal understanding of images and questions with relational reasoning leading to the correct answer. To bridge the semantic gap between these two modalities, previous works focus on the word-region alignments of all possible pairs without attending more attention to the corresponding word and object. Treating all pairs equally without consideration of relation consistency hinders the model’s performance. In this paper, to align the relation-consistent pairs and integrate the interpretability of VQA systems, we propose a Cross-modal Relational Reasoning Network (CRRN), to mask the inconsistent attention map and highlight the full latent alignments of corresponding word-region pairs. Specifically, we present two relational masks for inter-modal and intra-modal highlighting, inferring the more and less important words in sentences or regions in images. The attention interrelationship of consistent pairs can be enhanced with the shift of learning focus by masking the unaligned relations. Then, we propose two novel losses ℒCMAM and ℒSMAM with explicit supervision to capture the fine-grained interplay between vision and language. We have conduct thorough experiments to prove the effectiveness and achieve the competitive performance for reaching 61.74% on GQA benchmark.