Few-Shot Class-Incremental Learning on CUB200 2011 (test)
77.26Accuracy (Session 0)TEEN
Evaluation results
| Method | Links | |||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| TEENBackbone=ResNet-182023.12 | 77.26 | 76.13 | 72.81 | 68.16 | 67.77 | 64.4 | 63.25 | 62.29 | 61.19 | 60.32 | 59.31 | 18.13 | — | |
| FACTBackbone=ResNet-182023.12 | 75.9 | 73.23 | 70.84 | 66.13 | 65.56 | 62.15 | 61.74 | 59.83 | 58.41 | 57.89 | 56.94 | 18.96 | 0.83 | |
| CECBackbone=ResNet-182023.12 | 75.85 | 71.94 | 68.5 | 63.5 | 62.43 | 58.27 | 57.73 | 55.81 | 54.83 | 53.52 | 52.28 | 23.57 | 5.44 | |
| Decoupled-CosineBackbone=ResNet-182023.12 | 75.52 | 70.95 | 66.46 | 61.2 | 60.86 | 56.88 | 55.4 | 53.49 | 51.94 | 50.93 | 49.31 | 26.21 | 8.08 | |
| Decoupled-DeepEMDBackbone=ResNet-182023.12 | 75.35 | 70.69 | 66.68 | 62.34 | 59.76 | 56.54 | 54.61 | 52.52 | 50.73 | 49.2 | 47.6 | 27.75 | 9.62 | |
| Decoupled-NegCosineBackbone=ResNet-182023.12 | 74.96 | 70.57 | 66.62 | 61.32 | 60.09 | 56.06 | 55.03 | 52.78 | 51.5 | 50.08 | 48.47 | 26.49 | 7.36 | |
| iCaRLBackbone=ResNet-18, Exemplars=No2023.12 | 68.68 | 52.65 | 48.61 | 44.16 | 36.62 | 29.52 | 27.83 | 26.26 | 24.01 | 23.89 | 21.16 | 47.52 | 29.39 | |
| EEILBackbone=ResNet-18, Exemplars=No2023.12 | 68.68 | 53.63 | 47.91 | 44.2 | 36.3 | 27.46 | 25.93 | 24.7 | 23.95 | 24.13 | 22.11 | 46.57 | 28.44 | |
| RebalancingBackbone=ResNet-18, Exemplars=No2023.12 | 68.68 | 57.12 | 44.21 | 28.78 | 26.71 | 25.66 | 24.62 | 21.52 | 20.12 | 20.06 | 19.87 | 48.81 | 30.68 | |
| TOPICBackbone=ResNet-182023.12 | 68.68 | 62.49 | 54.81 | 49.99 | 45.25 | 41.4 | 38.35 | 35.36 | 32.22 | 28.31 | 26.26 | 42.4 | 24.27 |