5-way Few-shot Image Classification on FC-100 (test)
50.441-Shot Accuracy (5-way)Transductive fine-tuning
Evaluation results
| Method | Links | ||
|---|---|---|---|
| Transductive fine-tuningArchitecture=WRN-28-10, Pre-training Split=train + val2019.09 | 50.44 | 65.74 | |
| MetaOpt SVMArchitecture=ResNet-12*, Pre-training Split=train + val2019.09 | 47.2 | 62.5 | |
| Support-based initializationArchitecture=WRN-28-10, Pre-training Split=train + val2019.09 | 45.08 | 60.05 | |
| Fine-tuningArchitecture=WRN-28-10, Pre-training Split=train + val2019.09 | 43.8 | 64.4 | |
| Transductive fine-tuningArchitecture=WRN-28-10, Pre-training Split=train2019.09 | 43.16 | 57.57 | |
| MetaOpt SVMArchitecture=ResNet-12, Pre-training Split=train2019.09 | 41.1 | 55.5 | |
| TADAMArchitecture=ResNet-122019.09 | 40.1 | 56.1 | |
| Fine-tuningArchitecture=WRN-28-10, Pre-training Split=train2019.09 | 38.25 | 57.19 | |
| Support-based initializationArchitecture=WRN-28-10, Pre-training Split=train2019.09 | 36.82 | 49.72 |