Action Recognition In Videos On Something 3
Metrics
Top-1 Accuracy
Top-5 Accuracy
Results
Performance results of various models on this benchmark
| Paper Title | |||
|---|---|---|---|
| STM (16 frames, ImageNet pretraining) | 64.2 | 89.8 | STM: SpatioTemporal and Motion Encoding for Action Recognition |
| CPNet Res34, 5 CP | 57.65 | 83.95 | Learning Video Representations from Correspondence Proposals |
| 2-Stream TRN | 55.52 | 83.06 | Temporal Relational Reasoning in Videos |
| DIN | 34.11 | - | DenseImage Network: Video Spatial-Temporal Evolution Encoding and Understanding |
0 of 4 row(s) selected.