Video Classification on UCF-101 (Test Set Evaluation)
95.5AccuracyAMeFu-Net
Evaluation Results
| Method | Links | |||
|---|---|---|---|---|
| AMeFu-Netevaluation_protocol=5-way, shot=5-shot2020.10 | 95.5 | — | — | |
| AMeFu-Netevaluation_protocol=5-way, shot=3-shot2020.10 | 93.1 | — | — | |
| RGB Basenet++evaluation_protocol=5-way, shot=5-shot2020.10 | 92.9 | — | — | |
| RGB Basenetevaluation_protocol=5-way, shot=5-shot2020.10 | 92.1 | — | — | |
| RGB Basenet++evaluation_protocol=5-way, shot=3-shot2020.10 | 90 | — | — | |
| RGB Basenetevaluation_protocol=5-way, shot=3-shot2020.10 | 88.7 | — | — | |
| AMeFu-Netevaluation_protocol=5-way, shot=1-shot2020.10 | 85.1 | — | — | |
| ARNevaluation_protocol=5-way, shot=5-shot2020.10 | 83.1 | — | — | |
| RGB Basenet++evaluation_protocol=5-way, shot=1-shot2020.10 | 78.5 | — | — | |
| RGB Basenetevaluation_protocol=5-way, shot=1-shot2020.10 | 76.4 | — | — | |
| Supervised MAMLN-way=5, K-shot=1, Backbone=C3D, Initial Pre-training=Sports-1M, Meta-learning data=Kinetics2018.11 | 71.08 | 69.44 | — | |
| ARNevaluation_protocol=5-way, shot=1-shot2020.10 | 66.3 | — | — | |
| UMTRAN-way=5, K-shot=1, Backbone=C3D, Initial Pre-training=Sports-1M, Meta-learning data=unlabeled Kinetics2018.11 | 60.33 | 58.47 | — | |
| Pre-trained on KineticsN-way=5, K-shot=1, Backbone=C3D, Initial Pre-training=Sports-1M2018.11 | 45.51 | 42.49 | — | |
| Training from scratchN-way=5, K-shot=1, Backbone=C3D, Initial Pre-training=Sports-1M2018.11 | 29.3 | 20.48 | — | |
| I3DInput=32x22021.04 | — | — | 95.1 | |
| TEINetInput=16 (TSN)2021.04 | — | — | 96.7 | |
| TSMInput=8(TSN)2021.04 | — | — | 94.5 | |
| VidTr-LInput=32 x 22021.04 | — | — | 96.7 | |
| VidTr-MInput=16 x 42021.04 | — | — | 96.6 |