Video Question Answering on Env-QA (test)
44.05Attribute AccuracyMIST-CLIP
Evaluation Results
| Method | Links | ||||||
|---|---|---|---|---|---|---|---|
| MIST-CLIPfeature=CLIP (ViT-B/32)2022.12 | 44.05 | 58.13 | 42.54 | 56.83 | 40.32 | 48.97 | |
| MIST-AIOfeature=All-in-One-Base2022.12 | 43.63 | 55.17 | 40.99 | 55.44 | 39.54 | 47.19 | |
| TSEA2022.12 | 42.96 | 56.73 | 39.84 | 55.53 | 39.35 | 47.06 | |
| Temp[ATP]feature=CLIP (ViT-B/32)2022.12 | 42.87 | 53.49 | 38.35 | 55.25 | 38.65 | 45.43 | |
| AIOfeature=All-in-One-Base2022.12 | 41.78 | 52.98 | 37.57 | 55.16 | 38.5 | 44.86 | |
| ST-VQA2022.12 | 41.66 | 48.98 | 33.87 | 54.09 | 38.54 | 41.97 | |
| STAGE2022.12 | 39.49 | 49.93 | 34.52 | 55.32 | 37.98 | 42.53 | |
| CNN-LSTM2022.12 | 38.21 | 42.26 | 29.94 | 53.37 | 38.12 | 38.05 |