Few-shot image classification on CIFAR-FS 100 variant (test)
89.6AccuracyProtoNet+LLS
Evaluation Results
| Method | Links | |
|---|---|---|
| ProtoNet+LLSBackbone=ResNet-12, Shots=5-shot2021.11 | 89.6 | |
| MCT (Inst)Backbone=ResNet-12, Shots=5-shot2021.11 | 89.11 | |
| R2-D2+AugBackbone=ResNet-12, Shots=5-shot2021.11 | 87.63 | |
| S2M2_RBackbone=WRN-28-10, Shots=5-shot2021.11 | 87.47 | |
| ProtoNet* (baseline)Backbone=ResNet-12, Shots=5-shot2021.11 | 86.15 | |
| MetaOptNet-SVMBackbone=ResNet-12, Shots=5-shot2021.11 | 84.2 | |
| ProtoNet+LLSBackbone=ResNet-12, Shots=1-shot2021.11 | 78.76 | |
| MCT (Inst)Backbone=ResNet-12, Shots=1-shot2021.11 | 77.84 | |
| R2-D2+AugBackbone=ResNet-12, Shots=1-shot2021.11 | 76.51 | |
| S2M2_RBackbone=WRN-28-10, Shots=1-shot2021.11 | 74.81 | |
| MetaOptNet-SVMBackbone=ResNet-12, Shots=1-shot2021.11 | 72 | |
| ProtoNet* (baseline)Backbone=ResNet-12, Shots=1-shot2021.11 | 71.65 |