Audio-visual Classification on VGGSound Music
71.57Top-1 AccuracyAVAGENT
Evaluation Results
| Method | Links | |
|---|---|---|
| AVAGENTEvaluation Protocol=Fine-tuning2024.10 | 71.57 | |
| MAVILEvaluation Protocol=Fine-tuning2024.10 | 69.53 | |
| AV-MAEEvaluation Protocol=Fine-tuning2024.10 | 67.61 | |
| CAV-MAEEvaluation Protocol=Fine-tuning2024.10 | 67.26 | |
| AVAGENTEvaluation Protocol=Linear Probing2024.10 | 64.57 | |
| MAVILEvaluation Protocol=Linear Probing2024.10 | 61.95 | |
| AV-MAEEvaluation Protocol=Linear Probing2024.10 | 60.82 | |
| CAV-MAEEvaluation Protocol=Linear Probing2024.10 | 60.53 | |
| AudioMAEEvaluation Protocol=Fine-tuning2024.10 | 55.61 | |
| MAEEvaluation Protocol=Fine-tuning2024.10 | 52.39 | |
| AudioMAEEvaluation Protocol=Linear Probing2024.10 | 41.65 | |
| MAEEvaluation Protocol=Linear Probing2024.10 | 25.32 |