Image Classification on VOC07 (train)
88.9Top-1 AccuracySwAV
Evaluation results
| Method | Links | |
|---|---|---|
| SwAVarchitecture=R50-MLP (28M), pre-training epochs=800, evaluation augmentation=multi-crop, batch size=40962020.05 | 88.9 | |
| PCL v2architecture=R50-MLP (28M), pre-training epochs=2002020.05 | 85.4 | |
| MoCo v2architecture=R50-MLP (28M), pre-training epochs=2002020.05 | 84 | |
| PCLarchitecture=R50 (24M), pre-training epochs=2002020.05 | 82.3 | |
| PIRLarchitecture=R50 (24M), pre-training epochs=8002020.05 | 81.1 | |
| SimCLRarchitecture=R50-MLP (28M), pre-training epochs=1000, batch size=40962020.05 | 80.5 | |
| MoCoarchitecture=R50 (24M), pre-training epochs=2002020.05 | 79.2 | |
| DeepClusterarchitecture=VGG(15M), pre-training epochs=1002020.05 | 71.9 | |
| Jigsawarchitecture=R50 (24M), pre-training epochs=902020.05 | 64.5 | |
| Rotationarchitecture=R50 (24M)2020.05 | 63.9 |