| action-classification-on-kinetics-400 | VideoMAE (no extra data, ViT-H, 32x320x320) | #32 | Acc@1: 87.4Acc@5: 97.6 |
| action-classification-on-kinetics-400 | VideoMAE (no extra data, ViT-H) | #40 | Acc@1: 86.6Acc@5: 97.1 |
| action-classification-on-kinetics-400 | VideoMAE (no extra data, ViT-L, 32x320x320) | #44 | Acc@1: 86.1Acc@5: 97.3 |
| action-classification-on-kinetics-400 | VideoMAE (no extra data, ViT-L, 16x4) | #50 | Acc@1: 85.2Acc@5: 96.8 |
| action-classification-on-kinetics-400 | VideoMAE (no extra data, ViT-B, 16x4) | #74 | Acc@1: 81.5Acc@5: 95.1 |
| action-recognition-in-videos-on-something | VideoMAE (no extra data, ViT-L, 32x2) | #9 | Top-1 Accuracy: 75.4Top-5 Accuracy: 95.2Parameters: 305… |
| action-recognition-in-videos-on-something | VideoMAE (no extra data, ViT-L, 16frame) | #15 | Top-1 Accuracy: 74.3Top-5 Accuracy: 94.6Parameters: 305… |
| action-recognition-in-videos-on-something | VideoMAE (no extra data, ViT-B, 16frame) | #33 | Top-1 Accuracy: 70.8Top-5 Accuracy: 92.4Parameters: 87… |
| action-recognition-on-ava-v2-2 | VideoMAE (K400 pretrain+finetune, ViT-H, 16x4) | #9 | mAP: 39.5 |
| action-recognition-on-ava-v2-2 | VideoMAE (K700 pretrain+finetune, ViT-L, 16x4) | #10 | mAP: 39.3 |
| action-recognition-on-ava-v2-2 | VideoMAE (K400 pretrain+finetune, ViT-L, 16x4) | #12 | mAP: 37.8 |
| action-recognition-on-ava-v2-2 | VideoMAE (K400 pretrain, ViT-H, 16x4) | #14 | mAP: 36.5 |
| action-recognition-on-ava-v2-2 | VideoMAE (K700 pretrain, ViT-L, 16x4) | #15 | mAP: 36.1 |
| action-recognition-on-ava-v2-2 | VideoMAE (K400 pretrain, ViT-L, 16x4) | #18 | mAP: 34.3 |
| action-recognition-on-ava-v2-2 | VideoMAE (K400 pretrain+finetune, ViT-B, 16x4) | #21 | mAP: 31.8 |
| action-recognition-on-ava-v2-2 | VideoMAE (K400 pretrain, ViT-B, 16x4) | #31 | mAP: 26.7 |
| self-supervised-action-recognition-on-hmdb51 | VideoMAE | #5 | Top-1 Accuracy: 73.3Pre-Training Dataset: Kinetics400Frozen: false |
| self-supervised-action-recognition-on-hmdb51 | VideoMAE(no extra data) | #21 | Top-1 Accuracy: 62.6Pre-Training Dataset: no extra dataFrozen: false |
| self-supervised-action-recognition-on-ucf101 | VideoMAE | #5 | 3-fold Accuracy: 96.1Pre-Training Dataset: Kinetics400Frozen: false |
| self-supervised-action-recognition-on-ucf101 | VideoMAE(no extra data) | #19 | 3-fold Accuracy: 91.3Pre-Training Dataset: no extra dataFrozen: false |