Image Super Resolution On Set14 3X Upscaling
Métriques
PSNR
SSIM
Résultats
Résultats de performance de divers modèles sur ce benchmark
Tableau comparatif
| Nom du modèle | PSNR | SSIM |
|---|---|---|
| hmanet-hybrid-multi-axis-aggregation-network | 31.47 | 0.8585 |
| hierarchical-information-flow-for-generalized | 31.55 | 0.8616 |
| activating-more-pixels-in-image-super | 31.33 | 0.8576 |
| single-image-super-resolution-via-a-holistic | 30.79 | 0.8487 |
| a-framework-for-real-time-object-detection | 30.65 | 0.8493 |
| swinfir-revisiting-the-swinir-with-fast | 31.24 | 0.8566 |
| image-super-resolution-with-cross-scale-non | 30.66 | 0.8482 |
| ml-craist-multi-scale-low-high-frequency | 30.39 | 0.8488 |
| activating-more-pixels-in-image-super | 31.47 | 0.8584 |
| local-texture-estimator-for-implicit | 30.8 | - |
| channel-partitioned-windowed-attention-and | 31.15 | 0.8557 |
| image-restoration-using-convolutional-auto | 29.61 | 0.8341 |
| channel-partitioned-windowed-attention-and | 31.19 | 0.8559 |
| lcscnet-linear-compressing-based-skip | 29.87 | - |
| feedback-network-for-image-super-resolution | 30.1 | - |
| learning-deep-cnn-denoiser-prior-for-image | 27.72 | - |
| progressive-multi-scale-residual-network-for | 29.24 | 0.8087 |
| swinfir-revisiting-the-swinir-with-fast | 31.37 | - |
| pre-trained-image-processing-transformer | 30.85 | - |
| multi-level-wavelet-cnn-for-image-restoration | 30.16 | - |
| lightweight-image-super-resolution-with-1 | 30.32 | - |
| densely-residual-laplacian-super-resolution | 30.8 | 0.8498 |
| ml-craist-multi-scale-low-high-frequency | 30.23 | 0.8474 |
| beyond-a-gaussian-denoiser-residual-learning | 29.81 | - |