Image Enhancement On Mit Adobe 5K
Métriques
PSNR on proRGB
SSIM on proRGB
Résultats
Résultats de performance de divers modèles sur ce benchmark
Tableau comparatif
| Nom du modèle | PSNR on proRGB | SSIM on proRGB |
|---|---|---|
| rsfnet-a-white-box-image-retouching-approach | 25.49 | 0.924 |
| seplut-separable-image-adaptive-lookup-tables | 25.47 | 0.921 |
| adaint-learning-adaptive-intervals-for-3d | 25.49 | 0.926 |
| 4d-lut-learnable-context-aware-4d-lookup | 24.61 | 0.918 |
| pqdynamicisp-dynamically-controlled-image | 25.53 | 0.928 |
| difar-deep-image-formation-and-retouching | 24.2 | 0.88 |
| retinexformer-one-stage-retinex-based | 25.98 | 0.957 |
| deeplpf-deep-local-parametric-filters-for | 23.93 | 0.903 |
| multiple-transformation-function-estimation | 25.46 | - |
| learning-image-adaptive-3d-lookup-tables-for | 25.21 | 0.922 |