Fine-grained Image Classification on Cars
77.45Top-1 AccBYOL+CVSA
Evaluation results
| Method | Links | |
|---|---|---|
| BYOL+CVSABackbone=ResNet-50, Pre-training Setting=Second-stage only (800 epochs), Evaluation Protocol=Fine-tuned2021.06 | 77.45 | |
| InsLocBackbone=ResNet-50, Pre-training Setting=Second-stage only (800 epochs), Evaluation Protocol=Fine-tuned2021.06 | 76.61 | |
| BYOL+DiLoBackbone=ResNet-50, Pre-training Setting=Second-stage only (800 epochs), Evaluation Protocol=Fine-tuned2021.06 | 76.36 | |
| BYOLBackbone=ResNet-50, Pre-training Setting=Second-stage only (800 epochs), Evaluation Protocol=Fine-tuned2021.06 | 75.87 | |
| LooC*Backbone=ResNet-50, Pre-training Setting=Second-stage only (800 epochs), Evaluation Protocol=Fine-tuned2021.06 | 75.69 | |
| Rot-PredBackbone=ResNet-50, Pre-training Setting=Second-stage only (800 epochs), Evaluation Protocol=Fine-tuned2021.06 | 75.48 | |
| Rel-LocBackbone=ResNet-50, Pre-training Setting=Second-stage only (800 epochs), Evaluation Protocol=Fine-tuned2021.06 | 75.27 | |
| SimCLRBackbone=ResNet-50, Pre-training Setting=Second-stage only (800 epochs), Evaluation Protocol=Fine-tuned2021.06 | 75.12 | |
| MoCo.v2Backbone=ResNet-50, Pre-training Setting=Second-stage only (800 epochs), Evaluation Protocol=Fine-tuned2021.06 | 74.89 | |
| RandomBackbone=ResNet-50, Pre-training Setting=Random initialization, Evaluation Protocol=Fine-tuned2021.06 | 70.51 | |
| Fair Context Learning (FCL)mode=Test-Time Adaptation2026.02 | 68.58 | |
| MTAmode=Test-Time Adaptation2026.02 | 67.49 | |
| TPSmode=Test-Time Adaptation2026.02 | 67.21 | |
| ZEROmode=Test-Time Adaptation2026.02 | 67.03 | |
| R-TPTmode=Test-Time Adaptation2026.02 | 66.67 | |
| RLCFmode=Test-Time Adaptation2026.02 | 66.41 | |
| TTLmode=Test-Time Adaptation2026.02 | 66.25 | |
| TPTmode=Test-Time Adaptation2026.02 | 66.24 | |
| C-TPTmode=Test-Time Adaptation2026.02 | 65.71 | |
| CLIPmode=Zero-shot2026.02 | 65.51 |