Audio-Visual Action Recognition on HD-EPIC-SOUNDS
31.9Top-1 AccuracyTIM
Evaluation Results
| Method | Links | |
|---|---|---|
| TIMModality=A+V, Param.=435M, GFLOPs=460172025.03 | 31.9 | |
| CAVAModality=A+V, Param.=216M, GFLOPs=2362025.03 | 28.6 | |
| CA2STModality=A+V, Param.=320M, GFLOPs=4542025.03 | 28.1 | |
| Auditory SlowFastModality=A, Param.=26M, GFLOPs=42025.03 | 27.9 | |
| CASTModality=V, Param.=217M, GFLOPs=3912025.03 | 27.8 | |
| TIMModality=A, Param.=64M, GFLOPs=1162025.03 | 26.9 | |
| SSASTModality=A2025.03 | 25.1 |