Long-tailed Image Classification on CIFAR-10-LT IF=100 (test)
88.5Top-1 AccuracyGLMC + SAMN
Evaluation results
| Method | Links | |
|---|---|---|
| GLMC + SAMNBackbone=ResNet32, Imbalance Factor=1002026.06 | 88.5 | |
| GLMCBackbone=ResNet32, Imbalance Factor=1002026.06 | 88.2 | |
| SLAS + SAMNBackbone=ResNet32, Imbalance Factor=1002026.06 | 85.5 | |
| DiffuLTBackbone=ResNet32, Imbalance Factor=1002026.06 | 84.7 | |
| SLASBackbone=ResNet32, Imbalance Factor=1002026.06 | 84.2 | |
| CE + SAMNBackbone=ResNet32, Imbalance Factor=1002026.06 | 84.1 | |
| ADRWBackbone=ResNet32, Imbalance Factor=1002026.06 | 83.6 | |
| CSABackbone=ResNet32, Imbalance Factor=1002026.06 | 82.5 | |
| FBLBackbone Net=ResNet-32, IF=1002023.05 | 82.46 | |
| SAMBackbone=ResNet32, Imbalance Factor=1002026.06 | 81.9 | |
| LABackbone Net=ResNet-32, IF=1002023.05 | 80.92 | |
| CE (with WD in stage one)Backbone=ResNet32, Imbalance Factor=1002026.06 | 80 | |
| BBNBackbone Net=ResNet-32, IF=1002023.05 | 79.82 | |
| BBNBackbone=ResNet32, Imbalance Factor=1002026.06 | 79.8 | |
| LDAM-DRWBackbone Net=ResNet-32, IF=1002023.05 | 77.03 | |
| LDAM-DRWBackbone=ResNet32, Imbalance Factor=1002026.06 | 77 | |
| cRTBackbone=ResNet32, Imbalance Factor=1002026.06 | 75.7 | |
| CAM-BSBackbone=ResNet32, Imbalance Factor=1002026.06 | 75.4 | |
| CE lossBackbone Net=ResNet-32, IF=1002023.05 | 71.07 | |
| Focal LossBackbone=ResNet32, Imbalance Factor=1002026.06 | 70.4 |