Action Recognition on Diving-48 (val)
88.4AccuracyCAN
Evaluation results
| Method | Links | |
|---|---|---|
| CANBackbone=ResNet50, #Frame=162025.12 | 88.4 | |
| GC-TDNBackbone=ResNet50, #Frame=162025.12 | 87.6 | |
| GC-TSMBackbone=ResNet50, #Frame=162025.12 | 87.2 | |
| GC-TSNBackbone=ResNet50, #Frame=162025.12 | 86.8 | |
| TDNBackbone=ResNet50, #Frame=162025.12 | 84.6 | |
| TSMBackbone=ResNet50, #Frame=162025.12 | 83.2 | |
| GC-GSTBackbone=ResNet50, #Frame=162025.12 | 82.5 | |
| GENetBackbone=ResNet50, #Frame=162025.12 | 82.4 | |
| PMI-X3D-MBackbone=X3D-M, #Frame=162025.12 | 81.3 | |
| TimeSformer-LPre-training=ImageNet-1K2021.02 | 81 | |
| TimeSformer-LBackbone=Transformer, #Frame=962025.12 | 81 | |
| GSTBackbone=ResNet50, #Frame=162025.12 | 79.8 | |
| TSNBackbone=ResNet50, #Frame=162025.12 | 78.9 | |
| TimeSformer-HRPre-training=ImageNet-1K2021.02 | 78 | |
| TimeSformer-HRBackbone=Transformer, #Frame=162025.12 | 78 | |
| SlowFastPre-training=ImageNet-1K, Backbone=ResNet-101, Setting=16 x 82021.02 | 77.6 | |
| SlowFast,16x8Backbone=ResNet101, #Frame=64+162025.12 | 77.6 | |
| TimeSformerPre-training=ImageNet-1K2021.02 | 74.9 | |
| MG-X3D-MBackbone=X3D-M, #Frame=162025.12 | 74.6 | |
| X3D-MBackbone=X3D-M, #Frame=162025.12 | 73.5 |