Few-Shot Image Classification on Stanford Dogs (test)
67.561-shot AccuracyCausalFSFG
Evaluation results
| Method | Links | |||
|---|---|---|---|---|
| CausalFSFGBackbone=Conv-42025.12 | 67.56 | 82.83 | — | |
| C2-NetBackbone=Conv-4, Published In=AAAI 20242025.12 | 66.42 | 81.23 | — | |
| RCNBackbone=Res12, Type=Metric2020.09 | 66.24 | 81.5 | — | |
| Bi-FRNBackbone=Conv-4, Published In=AAAI 20232025.12 | 64.74 | 81.29 | — | |
| FRNBackbone=Conv-4, Published In=CVPR 20212025.12 | 60.41 | 79.26 | — | |
| CTXBackbone=Conv-4, Published In=NeurIPS 20202025.12 | 57.86 | 73.59 | — | |
| PARNBackbone=Conv-4, Published In=ICCV 20192025.12 | 55.86 | 68.06 | — | |
| RCNBackbone=Conv4-64, Type=Metric2020.09 | 54.29 | 72.65 | — | |
| FRN+TDMBackbone=Conv-4, Published In=CVPR 20222025.12 | 51.57 | 75.25 | — | |
| RelationBackbone=Conv-4, Published In=CVPR 20182025.12 | 47.35 | 66.2 | — | |
| GNNBackbone=Conv4-64, Type=Metric2020.09 | 46.98 | 62.27 | — | |
| DeepEMDBackbone=Conv-4, Published In=CVPR 20202025.12 | 46.73 | 65.74 | — | |
| ProtoNetBackbone=Conv-4, Published In=NeurIPS 20172025.12 | 46.66 | 70.77 | — | |
| DN4Backbone=Conv4-64, Type=Metric2020.09 | 45.73 | 66.33 | — | |
| LRPABNBackbone=Conv-4, Published In=TMM 20212025.12 | 45.72 | 60.94 | — | |
| SAMLBackbone=Conv-4, Published In=ICCV 20192025.12 | 45.46 | 59.65 | — | |
| MATCHING NETSBackbone=Conv4-64, Type=Metric2020.09 | 45.3 | 59.5 | — | |
| BSNet(D&C)Backbone=Conv-4, Published In=TIP 20212025.12 | 43.42 | 71.9 | — | |
| DN4Backbone=Conv-4, Published In=CVPR 20192025.12 | 39.08 | 69.81 | — | |
| PROTOTYPICAL NETSBackbone=Conv4-64, Type=Metric2020.09 | 37.59 | 48.19 | — | |
| PCMBackbone=Conv4-64, Type=Metric2020.09 | 28.78 | 46.92 | — | |
| BaselineBackbone=ResNet12, Shots=1, Ways=52020.10 | — | — | 63.53 | |
| BaselineBackbone=ResNet12, Shots=5, Ways=52020.10 | — | — | 79.95 | |
| Baseline++Backbone=ResNet12, Shots=1, Ways=52020.10 | — | — | 58.3 | |
| Baseline++Backbone=ResNet12, Shots=5, Ways=52020.10 | — | — | 73.77 | |
| Delta-encoderBackbone=ResNet12, Shots=1, Ways=52020.10 | — | — | 68.59 | |
| Delta-encoderBackbone=ResNet12, Shots=5, Ways=52020.10 | — | — | 78.6 | |
| MAMLBackbone=ResNet12, Shots=1, Ways=52020.10 | — | — | 66.56 | |
| MAMLBackbone=ResNet12, Shots=5, Ways=52020.10 | — | — | 79.32 | |
| MatchingNetBackbone=ResNet12, Shots=1, Ways=52020.10 | — | — | 65.87 | |
| MatchingNetBackbone=ResNet12, Shots=5, Ways=52020.10 | — | — | 80.7 | |
| MetaOptNetBackbone=ResNet12, Shots=1, Ways=52020.10 | — | — | 65.48 | |
| MetaOptNetBackbone=ResNet12, Shots=5, Ways=52020.10 | — | — | 79.39 | |
| MTLBackbone=ResNet12, Shots=1, Ways=52020.10 | — | — | 54.96 | |
| MTLBackbone=ResNet12, Shots=5, Ways=52020.10 | — | — | 68.76 | |
| ProtoNetBackbone=ResNet12, Shots=1, Ways=52020.10 | — | — | 65.02 | |
| ProtoNetBackbone=ResNet12, Shots=5, Ways=52020.10 | — | — | 83.69 | |
| RelationNetBackbone=ResNet12, Shots=1, Ways=52020.10 | — | — | 59.38 | |
| RelationNetBackbone=ResNet12, Shots=5, Ways=52020.10 | — | — | 79.1 | |
| Variational Feature DisentanglingBackbone=ResNet12, Shots=1, Ways=52020.10 | — | — | 76.24 | |
| Variational Feature DisentanglingBackbone=ResNet12, Shots=5, Ways=52020.10 | — | — | 88 |