Conditional Image Generation on Places365 5 classes 500 samples per class
71.1FIDcGANTransfer
Evaluation results
| Method | Links | ||
|---|---|---|---|
| cGANTransferBackbone=BigGAN, Pre-training dataset=ImageNet, KMMD Gaussian kernel sigma=1, Source=Current work2021.02 | 71.1 | 0.16 | |
| MineGANBackbone=BigGAN, Pre-training dataset=ImageNet, Source=adopted from [39]2021.02 | 82.3 | 0.47 | |
| BSABackbone=BigGAN, Pre-training dataset=ImageNet, KMMD Gaussian kernel sigma=1, Source=Current work evaluation2021.02 | 85.9 | 0.18 | |
| TransferGANBackbone=BigGAN, Pre-training dataset=ImageNet, Source=adopted from [39]2021.02 | 89.2 | 0.53 | |
| PPGNBackbone=BigGAN, Pre-training dataset=ImageNet, Source=adopted from [39]2021.02 | 139 | 0.56 | |
| ScratchBackbone=BigGAN, Pre-training dataset=None, Source=adopted from [39]2021.02 | 190 | 0.96 |