Object Classification on PACS ResNet-50 (test)
98.36Accuracy (Photo)MDGHybrid
Evaluation results
| Method | Links | |||||
|---|---|---|---|---|---|---|
| MDGHybridBackbone=ResNet-502020.06 | 98.36 | 86.74 | 82.32 | 82.66 | 87.52 | |
| MatchDGBackbone=ResNet-502020.06 | 97.94 | 85.61 | 82.12 | 78.76 | 86.11 | |
| RSCBackbone=ResNet-502020.06 | 97.92 | 87.89 | 82.16 | 83.35 | 87.83 | |
| RandMatchBackbone=ResNet-502020.06 | 97.89 | 82.16 | 81.68 | 80.45 | 85.54 | |
| DomainBedBackbone=ResNet-502020.06 | 97.8 | 88.1 | 77.9 | 79.1 | 85.7 | |
| CORALBackbone=ResNet-502020.06 | 97.6 | 87.7 | 79.2 | 79.4 | 86 | |
| IRMBackbone=ResNet-502020.06 | 96.7 | 85 | 77.6 | 78.5 | 84.4 |