Visual Question Answering (Multiple-Choice) on VQA 1.0 (test-standard)
69.97Accuracy (All)Dual-MFA
Evaluation Results
| Method | Links | ||||
|---|---|---|---|---|---|
| Dual-MFA2017.11 | 69.97 | — | — | — | |
| DANBackbone=ResNet2016.11 | 69 | — | — | — | |
| MLB2017.11 | 68.89 | — | — | — | |
| VQA-Machine2017.11 | 67.8 | — | — | — | |
| Ours-ResNetBackbone=ResNet, type=Single Model2016.06 | 67.3 | 81.7 | 40 | 61 | |
| RAUBackbone=ResNet2016.11 | 67.3 | — | — | — | |
| MRN2017.11 | 66.33 | — | — | — | |
| (Kim et al. 2016) ResNet*Backbone=ResNet, type=Single Model2016.06 | 66.3 | 82.4 | 39.6 | 58.4 | |
| MRNBackbone=ResNet2016.11 | 66.3 | — | — | — | |
| (Lu et al. 2016) ResNet*Backbone=ResNet, type=Single Model2016.06 | 66.1 | — | — | — | |
| HieCoAttBackbone=ResNet2016.11 | 66.1 | — | — | — | |
| HieCoAtt2017.11 | 66.07 | — | — | — | |
| QRU2017.11 | 65.43 | — | — | — | |
| FDA2017.11 | 64.18 | — | — | — | |
| (Antol et al. 2015)type=Single Model2016.06 | 63.1 | 80.6 | 37.7 | 53.6 | |
| VQA team2016.11 | 63.1 | — | — | — | |
| (Noh, Seo, and Han 2016)type=Single Model2016.06 | 62.7 | 80.4 | 38.8 | 52.8 | |
| DPPnet2016.11 | 62.7 | — | — | — | |
| DPPnet2017.11 | 62.69 | — | — | — | |
| DPPnet2015.12 | 62.69 | 80.35 | 38.79 | 52.79 | |
| Region Sel.2017.11 | 62.43 | — | — | — | |
| (Shih, Singh, and Hoiem 2016)type=Single Model2016.06 | 62.4 | 77.2 | 33.5 | 56.1 | |
| (Zhou et al. 2015)type=Single Model2016.06 | 62 | 76.9 | 37.3 | 54.6 | |
| iBOWIMG2016.11 | 62 | — | — | — | |
| iBOWING2017.11 | 61.97 | — | — | — | |
| iBOWIMG2015.12 | 61.97 | 76.86 | 37.3 | 54.6 | |
| LSTM Q+I2017.11 | 57.57 | — | — | — |