Visual Question Answering On Coco Visual 1
평가 지표
Percentage correct
평가 결과
이 벤치마크에서 각 모델의 성능 결과
모델 이름 | Percentage correct | Paper Title | Repository |
|---|---|---|---|
| MCB 7 att. | 70.1 | Multimodal Compact Bilinear Pooling for Visual Question Answering and Visual Grounding | |
| 3-Modalities: Unary + Pairwise + Ternary (ResNet) | 69.3 | High-Order Attention Models for Visual Question Answering | |
| Dual-MFA | 70.04 | Co-attending Free-form Regions and Detections with Multi-modal Multiplicative Feature Embedding for Visual Question Answering | |
| HQI+ResNet | 66.1 | Hierarchical Question-Image Co-Attention for Visual Question Answering | |
| MRN | 66.3 | Multimodal Residual Learning for Visual QA | |
| iBOWIMG baseline | 62.0 | Simple Baseline for Visual Question Answering | |
| RelAtt | 69.60 | R-VQA: Learning Visual Relation Facts with Semantic Attention for Visual Question Answering | |
| joint-loss | 67.3 | Training Recurrent Answering Units with Joint Loss Minimization for VQA | - |
| LSTM Q+I | 63.1 | VQA: Visual Question Answering | |
| FDA | 64.2 | A Focused Dynamic Attention Model for Visual Question Answering | - |
0 of 10 row(s) selected.