| action-classification-on-kinetics-400 | ViViT-H/16x2 (JFT) | #53 | Acc@1: 84.9Acc@5: 95.8 |
| action-classification-on-kinetics-400 | ViViT-L/16x2 320 | #175 | Acc@5: 94.7 |
| action-classification-on-kinetics-600 | ViViT-H/16x2 (JFT) | #23 | Top-1 Accuracy: 85.8Top-5 Accuracy: 96.5 |
| action-classification-on-kinetics-600 | ViViT-L/16x2 | #28 | Top-1 Accuracy: 84.3Top-5 Accuracy: 95.6 |
| action-classification-on-kinetics-600 | ViViT-L/16x2 (320x320) | #35 | Top-1 Accuracy: 83.0Top-5 Accuracy: 95.7 |
| action-classification-on-moments-in-time | ViViT-L/16x2 | #24 | Top 5 Accuracy: 64.9 |
| action-recognition-in-videos-on-something | ViViT-L/16x2 Fact. encoder | #76 | Top-1 Accuracy: 65.4Top-5 Accuracy: 89.8 |
| action-recognition-on-epic-kitchens-100 | ViViT-L/16x2 Fact. encoder | #19 | Action@1: 44.0Verb@1: 66.4Noun@1: 56.8 |