Metric: Accuracy (higher is better)
| # | Model↕ | Accuracy▼ | Augmentations | Paper | Date↕ | Code |
|---|---|---|---|---|---|---|
| 1 | MoCo-v2 (ResNet-50) | 88.03 | Yes | - | - | Code |
| 2 | MoCo-v2 (ResNet-50) | 85.88 | Yes | Benchmarking Self-Supervised Learning on Diverse... | 2022-12-09 | Code |
| 3 | Barlow Twins (ResNet-50) | 84.03 | Yes | - | - | Code |
| 4 | SwAV (ResNet-50) | 83.21 | Yes | - | - | Code |
| 5 | Supervised (ViT-S/16) | 81.68 | Yes | Benchmarking Self-Supervised Learning on Diverse... | 2022-12-09 | Code |
| 6 | Barlow Rwins (ResNet-50) | 81.27 | Yes | Benchmarking Self-Supervised Learning on Diverse... | 2022-12-09 | Code |
| 7 | DINO (ViT-S/16) | 79.43 | Yes | Benchmarking Self-Supervised Learning on Diverse... | 2022-12-09 | Code |
| 8 | Supervised (ResNet-50) | 78.92 | Yes | Benchmarking Self-Supervised Learning on Diverse... | 2022-12-09 | Code |
| 9 | SwAV (ResNet-50) | 77.99 | Yes | Benchmarking Self-Supervised Learning on Diverse... | 2022-12-09 | Code |