Few-shot classification on NAB (test)
95.22AccuracyVariational Feature Disentangling
Evaluation results
| Method | Links | |
|---|---|---|
| Variational Feature DisentanglingBackbone=ResNet12, Shots=5, Ways=52020.10 | 95.22 | |
| MetaOptNetBackbone=ResNet12, Shots=5, Ways=52020.10 | 93.31 | |
| Delta-encoderBackbone=ResNet12, Shots=5, Ways=52020.10 | 92.32 | |
| Baseline++Backbone=ResNet12, Shots=5, Ways=52020.10 | 90.99 | |
| ProtoNetBackbone=ResNet12, Shots=5, Ways=52020.10 | 89.72 | |
| MAMLBackbone=ResNet12, Shots=5, Ways=52020.10 | 88.87 | |
| Variational Feature DisentanglingBackbone=ResNet12, Shots=1, Ways=52020.10 | 88.62 | |
| MatchingNetBackbone=ResNet12, Shots=5, Ways=52020.10 | 88.17 | |
| BaselineBackbone=ResNet12, Shots=5, Ways=52020.10 | 87.91 | |
| MTLBackbone=ResNet12, Shots=5, Ways=52020.10 | 87.74 | |
| RelationNetBackbone=ResNet12, Shots=5, Ways=52020.10 | 85.47 | |
| MetaOptNetBackbone=ResNet12, Shots=1, Ways=52020.10 | 84.56 | |
| MAMLBackbone=ResNet12, Shots=1, Ways=52020.10 | 80.08 | |
| Delta-encoderBackbone=ResNet12, Shots=1, Ways=52020.10 | 79.42 | |
| MTLBackbone=ResNet12, Shots=1, Ways=52020.10 | 78.69 | |
| Baseline++Backbone=ResNet12, Shots=1, Ways=52020.10 | 76 | |
| MatchingNetBackbone=ResNet12, Shots=1, Ways=52020.10 | 73.91 | |
| ProtoNetBackbone=ResNet12, Shots=1, Ways=52020.10 | 73.6 | |
| BaselineBackbone=ResNet12, Shots=1, Ways=52020.10 | 70.36 | |
| RelationNetBackbone=ResNet12, Shots=1, Ways=52020.10 | 67.41 |