Audio-to-Visual Knowledge Distillation on AVE
34.16AccuracyUCMKD
Evaluation results
| Method | Links | |
|---|---|---|
| UCMKDBackbone=ResNet-18, Batch size=64, Epoch=100, Optimizer=SGD, Learning rate=1e-2, Unpaired cross-modal setting=true2026.06 | 34.16 | |
| NORMBackbone=ResNet-18, Batch size=64, Epoch=100, Optimizer=SGD, Learning rate=1e-2, Unpaired cross-modal setting=true2026.06 | 27.68 | |
| REVIEWBackbone=ResNet-18, Batch size=64, Epoch=100, Optimizer=SGD, Learning rate=1e-2, Unpaired cross-modal setting=true2026.06 | 27.12 |