Image Super Resolution On Ffhq 1024 X 1024 4X
Métriques
FID
MS-SSIM
PSNR
SSIM
Résultats
Résultats de performance de divers modèles sur ce benchmark
Nom du modèle | FID | MS-SSIM | PSNR | SSIM | Paper Title | Repository |
|---|---|---|---|---|---|---|
| EnhanceNet | 19.07 | 0.934 | 29.42 | 0.832 | EnhanceNet: Single Image Super-Resolution Through Automated Texture Synthesis | |
| FSRCNN | 23.97 | 0.951 | 24.71 | 0.804 | Accelerating the Super-Resolution Convolutional Neural Network | |
| SRCNN | 31.84 | 0.924 | 27.40 | 0.801 | Image Super-Resolution Using Deep Convolutional Networks | |
| SRFBN | 17.14 | 0.931 | 27.90 | 0.822 | Feedback Network for Image Super-Resolution | |
| EDSR | 15.54 | 0.933 | 28.34 | 0.827 | Enhanced Deep Residual Networks for Single Image Super-Resolution | |
| SRGAN | 60.67 | 0.807 | 21.49 | 0.515 | Photo-Realistic Single Image Super-Resolution Using a Generative Adversarial Network | |
| ESRGAN | 72.73 | 0.782 | 19.84 | 0.353 | ESRGAN: Enhanced Super-Resolution Generative Adversarial Networks | |
| HiFaceGAN | 1.978 | 0.975 | 33.04 | 0.875 | HiFaceGAN: Face Renovation via Collaborative Suppression and Replenishment | |
| CAGFace | 12.4 | 0.971 | 34.1 | 0.906 | Component Attention Guided Face Super-Resolution Network: CAGFace |
0 of 9 row(s) selected.