Fine-grained classification on Flowers-102
79.8cACCFiNDR
Evaluation results
| Method | Links | ||
|---|---|---|---|
| FiNDRvocabulary_type=Automatically discovered, label_discovery_set=3 images per class, average_runs=102025.12 | 79.8 | 56.5 | |
| CLIP (zero-shot)protocol=zero-shot, ground_truth_class_names=known2025.12 | 69.7 | 77.8 | |
| CaSEDvocabulary_type=Pre-defined2025.12 | 67.2 | 52.3 | |
| KMeansvocabulary_type=Index-based, without_vocabulary=true2025.12 | 66.9 | — | |
| E-FineRvocabulary_type=Automatically discovered, label_discovery_set=3 images per class, average_runs=102025.12 | 64.8 | 54 | |
| FineRvocabulary_type=Automatically discovered, label_discovery_set=3 images per class, average_runs=102025.12 | 63.8 | 51.3 | |
| BLIP-2vocabulary_type=Automatically discovered, label_discovery_set=3 images per class, average_runs=102025.12 | 61.9 | 59.1 | |
| WordNetvocabulary_type=Pre-defined2025.12 | 42.1 | 49.8 | |
| CLIP-Sinkhornvocabulary_type=Index-based, without_vocabulary=true2025.12 | 30.9 | — | |
| DINO-Sinkhornvocabulary_type=Index-based, without_vocabulary=true2025.12 | 17.9 | — | |
| CLEVERvocabulary_type=Pre-defined2025.12 | 6.2 | — |