BLOCK: Bilinear Superdiagonal Fusion for Visual Question Answering and Visual Relationship Detection

Benchmark Model Rank Results
visual-question-answering-on-vqa-v2-test-devBLOCK#35Accuracy: 67.58
visual-question-answering-on-vqa-v2-test-stdBLOCK#29overall: 67.9
visual-relationship-detection-on-vrdBLOCK#2R@100: 92.58R@50: 86.58
visual-relationship-detection-on-vrd-1BLOCK#2R@100: 20.96R@50: 19.06
visual-relationship-detection-on-vrd-phraseBLOCK#1R@100: 28.96R@50: 26.32