Image Super Resolution On Ffhq 1024 X 1024 4X
Metrics
FID
MS-SSIM
PSNR
SSIM
Results
Performance results of various models on this benchmark
Comparison Table
| Model Name | FID | MS-SSIM | PSNR | SSIM |
|---|---|---|---|---|
| enhancenet-single-image-super-resolution | 19.07 | 0.934 | 29.42 | 0.832 |
| accelerating-the-super-resolution | 23.97 | 0.951 | 24.71 | 0.804 |
| image-super-resolution-using-deep | 31.84 | 0.924 | 27.40 | 0.801 |
| feedback-network-for-image-super-resolution | 17.14 | 0.931 | 27.90 | 0.822 |
| enhanced-deep-residual-networks-for-single | 15.54 | 0.933 | 28.34 | 0.827 |
| photo-realistic-single-image-super-resolution | 60.67 | 0.807 | 21.49 | 0.515 |
| esrgan-enhanced-super-resolution-generative | 72.73 | 0.782 | 19.84 | 0.353 |
| hifacegan-face-renovation-via-collaborative | 1.978 | 0.975 | 33.04 | 0.875 |
| component-attention-guided-face-super | 12.4 | 0.971 | 34.1 | 0.906 |