Image Generation on Oxford Flowers (FID)
12.64FIDDiff-Tuning
Evaluation results
| Method | Links | ||
|---|---|---|---|
| Diff-Tuning2024.06 | 12.64 | 24.2 | |
| DiffFitParameter-efficient=true2024.06 | 14.02 | — | |
| Full Fine-tuning2024.06 | 16.67 | — | |
| SADM (full)Pre-training=ImageNet, Resolution=256x256, Tuning Strategy=Full Fine-tuning2024.02 | 18.18 | — | |
| SADM (parameter-efficient)Pre-training=ImageNet, Resolution=256x256, Tuning Strategy=Parameter-efficient2024.02 | 19.37 | — | |
| DiffFitPre-training=ImageNet, Resolution=256x256, Tuning Strategy=Parameter-efficient2024.02 | 20.18 | — | |
| BitFitPre-training=ImageNet, Resolution=256x256, Tuning Strategy=Parameter-efficient2024.02 | 20.31 | — | |
| BitFitParameter-efficient=true, Source=Reported from [53]2024.06 | 20.31 | — | |
| Full Fine-tuning with DDPMPre-training=ImageNet, Resolution=256x256, Tuning Strategy=Full Fine-tuning2024.02 | 21.05 | — | |
| AdaptFormerPre-training=ImageNet, Resolution=256x256, Tuning Strategy=Parameter-efficient2024.02 | 21.24 | — | |
| AdaptFormerParameter-efficient=true, Source=Reported from [53]2024.06 | 21.36 | — | |
| VPTPre-training=ImageNet, Resolution=256x256, Tuning Strategy=Parameter-efficient2024.02 | 25.59 | — | |
| VPT-DeepParameter-efficient=true, Source=Reported from [53]2024.06 | 25.59 | — | |
| LoRAPre-training=ImageNet, Resolution=256x256, Tuning Strategy=Parameter-efficient2024.02 | 164.13 | — | |
| LoRAParameter-efficient=true, Source=Reported from [53]2024.06 | 164.13 | — |