Image Captioning on COCO Karpathy 5000 (test)
71.8BLEU-1Hard-Attention
Evaluation Results
| Method | Links | |||||||
|---|---|---|---|---|---|---|---|---|
| Hard-Attention2017.03 | 71.8 | 50.4 | 35.7 | 25 | 23 | — | — | |
| Image + ActionsimSitu features=Actions2017.03 | 71.5 | 54.6 | 40.9 | 30.9 | 24.7 | 94.5 | 17.6 | |
| Image + NounsimSitu features=Nouns2017.03 | 71.5 | 54.6 | 41.1 | 31.1 | 24.8 | 95.2 | 17.7 | |
| ATT-FCNEnsemble=true2017.03 | 70.9 | 53.7 | 40.2 | 30.4 | 24.3 | — | — | |
| NeuralTalk2Note=Modified baseline2017.03 | 70.8 | 53.7 | 40.1 | 30.1 | 24.5 | 93 | 17.3 | |
| Soft-Attention2017.03 | 70.7 | 49.2 | 34.4 | 24.3 | 23.9 | — | — | |
| img-gLSTMvariant=v22017.03 | 67 | 49.1 | 35.8 | 26.4 | 22.7 | 81.3 | — | |
| img-gLSTMvariant=v12017.03 | 64.7 | 45.9 | 31.1 | 21.4 | 20.4 | 67.7 | — | |
| LRCN2017.03 | 62.8 | 44.2 | 30.4 | 21 | — | — | — |