Image Classification on FGVC (Top-5 Accuracy)
95.23Top-5 AccuracyHACE
Evaluation results
| Method | Links | |
|---|---|---|
| HACEArchitecture=ResNet-50, Training Protocol=end-to-end training2026.05 | 95.23 | |
| HACEArchitecture=ResNet-18, Training Protocol=end-to-end training2026.05 | 94.15 | |
| HACEArchitecture=ResNet-34, Training Protocol=end-to-end training2026.05 | 93.97 | |
| SCEArchitecture=ResNet-34, Training Protocol=end-to-end training2026.05 | 93.64 | |
| SCEArchitecture=ResNet-50, Training Protocol=end-to-end training2026.05 | 90.58 | |
| SCEArchitecture=ResNet-18, Training Protocol=end-to-end training2026.05 | 88.63 | |
| HACEArchitecture=Swin-T, Training Protocol=end-to-end training2026.05 | 73.06 | |
| HACEArchitecture=ConvNeXt-T, Training Protocol=end-to-end training2026.05 | 66.34 | |
| SCEArchitecture=Swin-T, Training Protocol=end-to-end training2026.05 | 66.31 | |
| SCEArchitecture=ConvNeXt-T, Training Protocol=end-to-end training2026.05 | 57.01 | |
| HACEArchitecture=ViT-B/16, Training Protocol=end-to-end training2026.05 | 45.21 | |
| SCEArchitecture=ViT-B/16, Training Protocol=end-to-end training2026.05 | 33.99 |