Fine-grained Image Classification on Flowers (test)
97.9AccuracySR-GNN
Evaluation Results
| Method | Links | |
|---|---|---|
| SR-GNN2022.09 | 97.9 | |
| SotA2022.09 | 97.7 | |
| GIF-SDBackbone=ResNet-50, Expansion Ratio=20x2022.11 | 88.3 | |
| GIF-DALLEBackbone=ResNet-50, Expansion Ratio=20x2022.11 | 88.2 | |
| GIF-MAEBackbone=ResNet-50, Expansion Ratio=20x2022.11 | 84.4 | |
| DALL-E2Backbone=ResNet-50, Expansion Ratio=20x2022.11 | 84.1 | |
| RandAugmentBackbone=ResNet-50, Expansion Ratio=20x2022.11 | 83.8 | |
| GridMaskBackbone=ResNet-50, Expansion Ratio=20x2022.11 | 80.7 | |
| SDBackbone=ResNet-50, Expansion Ratio=20x2022.11 | 78.8 | |
| CutoutBackbone=ResNet-50, Expansion Ratio=20x2022.11 | 77.8 | |
| MAEBackbone=ResNet-50, Expansion Ratio=20x2022.11 | 76.3 | |
| Distillation of CLIPBackbone=ResNet-50, Expansion Ratio=1x2022.11 | 75.1 | |
| OriginalBackbone=ResNet-50, Expansion Ratio=1x2022.11 | 74.1 | |
| CLIPBackbone=CLIP, Expansion Ratio=1x, Evaluation Protocol=Zero-shot2022.11 | 65.9 |