Visual Question Answering On Coco Visual 4
Metrics
Percentage correct
Results
Performance results of various models on this benchmark
Comparison Table
| Model Name | Percentage correct |
|---|---|
| vqa-visual-question-answering | 58.2 |
| training-recurrent-answering-units-with-joint | 63.2 |
| r-vqa-learning-visual-relation-facts-with | 65.69 |
| co-attending-free-form-regions-and-detections | 66.09 |
| simple-baseline-for-visual-question-answering | 55.9 |
| image-captioning-and-visual-question | 59.5 |
| multimodal-residual-learning-for-visual-qa | 61.8 |
| a-focused-dynamic-attention-model-for-visual | 59.5 |
| hierarchical-question-image-co-attention-for | 62.1 |
| ask-attend-and-answer-exploring-question | 58.2 |
| multimodal-compact-bilinear-pooling-for | 66.5 |
| dynamic-memory-networks-for-visual-and | 60.4 |
| question-guided-hybrid-convolution-for-visual | 65.90 |
| stacked-attention-networks-for-image-question | 58.9 |