Visual Prompt Tuning On Vtab 1K Natural 7
评估指标
Mean Accuracy
评测结果
各个模型在此基准测试上的表现结果
模型名称 | Mean Accuracy | Paper Title | Repository |
|---|---|---|---|
| SPT-Shallow(ViT-B/16_MAE_pretrained_ImageNet-1K) | 62.53 | Revisiting the Power of Prompt for Visual Tuning | - |
| VPT-Shallow(ViT-B/16_MoCo_v3_pretrained_ImageNet-1K) | 67.34 | Visual Prompt Tuning | |
| SPT-Deep(ViT-B/16_MoCo_v3_pretrained_ImageNet-1K) | 76.20 | Revisiting the Power of Prompt for Visual Tuning | - |
| VPT-Shallow(ViT-B/16_MAE_pretrained_ImageNet-1K) | 39.96 | Visual Prompt Tuning | |
| SPT-Shallow(ViT-B/16_MoCo_v3_pretrained_ImageNet-1K) | 74.47 | Revisiting the Power of Prompt for Visual Tuning | - |
| SPT-Deep(ViT-B/16_MAE_pretrained_ImageNet-1K) | 67.19 | Revisiting the Power of Prompt for Visual Tuning | - |
| VPT-Deep(ViT-B/16_MoCo_v3_pretrained_ImageNet-1K) | 70.27 | Visual Prompt Tuning | |
| GateVPT(ViT-B/16_MoCo_v3_pretrained_ImageNet-1K) | 74.84 | Improving Visual Prompt Tuning for Self-supervised Vision Transformers | |
| VPT-Deep(ViT-B/16_MAE_pretrained_ImageNet-1K) | 36.02 | Visual Prompt Tuning | |
| GateVPT(ViT-B/16_MAE_pretrained_ImageNet-1K) | 47.61 | Improving Visual Prompt Tuning for Self-supervised Vision Transformers |
0 of 10 row(s) selected.