R-VQA: Learning Visual Relation Facts with Semantic Attention for Visual Question Answering

Open in new window