Image Classification on PASCAL VOC 2007
92.3AccuracyxCLIP
Evaluation Results
| Method | Links | |
|---|---|---|
| xCLIPLinear probing=true, Backbone=ViT-B/16, Pre-train=IT35M2022.10 | 92.3 | |
| CLIPLinear probing=true, Backbone=ViT-B/16, Pre-train=IT35M2022.10 | 91.1 | |
| nCLIPLinear probing=true, Backbone=ViT-B/16, Pre-train=IT35M2022.10 | 90.1 | |
| LogMETop-k=k=32022.07 | 86.89 | |
| NLEEPTop-k=k=32022.07 | 86.89 | |
| SFDATop-k=k=32022.07 | 86.89 | |
| SFDAcomTop-k=k=32022.07 | 86.89 | |
| NLEEPTop-k=k=22022.07 | 86.64 | |
| TWISTEvaluation Protocol=Fine-tune, Backbone=ResNet-50, Pre-trained=ImageNet2021.10 | 86.5 | |
| LogMETop-k=k=22022.07 | 86.46 | |
| SFDATop-k=k=22022.07 | 86.46 | |
| SFDAcomTop-k=k=22022.07 | 86.46 | |
| TWISTEvaluation Protocol=Linear evaluation, Backbone=ResNet-50, Pre-trained=ImageNet2021.10 | 85.7 | |
| EVA-02-CLIP-E/14+zero-shot evaluation protocol=true2023.03 | 85.6 | |
| BYOLEvaluation Protocol=Fine-tune, Backbone=ResNet-50, Pre-trained=ImageNet2021.10 | 85.4 | |
| SupervisedEvaluation protocol=Fine-tuned, Pre-training=ImageNet, Backbone=ResNet-502020.02 | 85 | |
| SUPEvaluation Protocol=Fine-tune, Backbone=ResNet-50, Pre-trained=ImageNet2021.10 | 85 | |
| SimCLREvaluation Protocol=Fine-tune, Backbone=ResNet-50, Pre-trained=ImageNet2021.10 | 84.5 | |
| SimCLREvaluation protocol=Fine-tuned, Pre-training=ImageNet, Backbone=ResNet-502020.02 | 84.1 | |
| SupervisedEvaluation protocol=Linear evaluation, Pre-training=ImageNet, Backbone=ResNet-502020.02 | 82.8 | |
| SUPEvaluation Protocol=Linear evaluation, Backbone=ResNet-50, Pre-trained=ImageNet2021.10 | 82.8 | |
| BYOLEvaluation Protocol=Linear evaluation, Backbone=ResNet-50, Pre-trained=ImageNet2021.10 | 82.5 | |
| SimCLREvaluation protocol=Linear evaluation, Pre-training=ImageNet, Backbone=ResNet-502020.02 | 80.5 | |
| SimCLREvaluation Protocol=Linear evaluation, Backbone=ResNet-50, Pre-trained=ImageNet2021.10 | 80.5 | |
| EVA-02-CLIP-B/16zero-shot evaluation protocol=true2023.03 | 80.2 | |
| OpenAI CLIP-B/16zero-shot evaluation protocol=true2023.03 | 78.3 | |
| Random initEvaluation protocol=Fine-tuned, Backbone=ResNet-502020.02 | 67.3 | |
| RandomEvaluation Protocol=Fine-tune, Backbone=ResNet-50, Pre-trained=ImageNet2021.10 | 67.3 |