Pretrained models which serve as a showcase of traiNNer-redux. These models were trained on a single RTX 4090 and surpass the metrics of paper models trained on 4 A100 GPUs. These models were trained following the methods outlined in the training guidelines.
DAT2
| Model |
Urban100 PSNR |
Urban100 SSIM |
| DAT-2 x2 |
34.31 |
0.9457 |
| 2x_DF2K_Redux_DAT2 (Ours) |
34.33 |
0.9471 |
| DAT-2 x4 |
27.86 |
0.8341 |
| 4x_DF2K_Redux_DAT2 (Ours) |
27.86 |
0.8372 |
PLKSR
| Model |
Urban100 PSNR |
Urban100 SSIM |
| PLKSR DF2K x2 |
33.36 |
0.9395 |
| 2x_DF2K_Redux_RealPLKSRLayerNorm (Ours) |
33.44 |
0.9412 |
| PLKSR DF2K x4 |
26.85 |
0.8097 |
| 4x_DF2K_Redux_RealPLKSRLayerNorm (Ours) |
26.94 |
0.8140 |
MoESR
| Model |
Urban100 PSNR |
Urban100 SSIM |
| 4x_DF2K_MoESR_500k |
27.05 |
0.8177 |
RCAN
| Model |
Urban100 PSNR |
Urban100 SSIM |
| RCAN_BIX2 |
33.34 |
0.9384 |
| RCANit_x2 |
33.62 |
0.9410 |
| 2x_DF2K_Redux_RCAN (Ours) |
33.65 |
0.9426 |
| RCAN_BIX4 |
26.82 |
0.8087 |
| RCANit_x4 |
27.16 |
0.8168 |
| 4x_DF2K_Redux_RCAN (Ours) |
27.18 |
0.8202 |
Compact
| Model |
Urban100 PSNR |
Urban100 SSIM |
| 2x_DF2K_Redux_Compact |
31.72 |
0.9257 |
ArtCNN
| Model |
Urban100 PSNR |
Urban100 SSIM |
| 2x_DF2K_ArtCNN_R8F48 |
31.82 |
0.9266 |
| 2x_DF2K_ArtCNN_R8F64 |
32.10 |
0.9293 |
| 2x_DF2K_ArtCNN_R16F96 |
32.81 |
0.9358 |
