Image Classification on Caltech Camera Traps (CCT) (test)
61.84AccuracyOATTA
Evaluation results
| Method | Links | |
|---|---|---|
| OATTABackbone=ConvNeXt2026.01 | 61.84 | |
| OATTA + TTAUGBackbone=ResNet502026.01 | 59.46 | |
| Source-OnlyBackbone=ConvNeXt2026.01 | 58.88 | |
| OATTA + MC-DRBackbone=ResNet502026.01 | 57.09 | |
| OATTABackbone=ResNet502026.01 | 57.08 | |
| TTAUGBackbone=ResNet502026.01 | 56.95 | |
| OATTA + MEMOBackbone=ResNet502026.01 | 56.19 | |
| OATTABackbone=ViT Base2026.01 | 55.24 | |
| MEMOBackbone=ResNet502026.01 | 54.49 | |
| MC-DRBackbone=ResNet502026.01 | 54.32 | |
| Source-OnlyBackbone=ResNet502026.01 | 54.3 | |
| Source-OnlyBackbone=ViT Base2026.01 | 53.17 | |
| TentBackbone=ResNet502026.01 | 52.87 | |
| OATTA + TentBackbone=ResNet502026.01 | 52.29 |