Image Denoising on Monte Carlo rendered image 64 spp
28.25PSNROurs
Evaluation results
| Method | Links | |||
|---|---|---|---|---|
| OursTraining noise protocol=Gaussian noise, σ = 15, Training strategy=masked training2023.03 | 28.25 | 76.94 | 0.3348 | |
| DnCNNTraining noise protocol=Gaussian noise, σ = 152023.03 | 26.28 | 67.79 | 0.4216 | |
| RIDNetTraining noise protocol=Gaussian noise, σ = 152023.03 | 26.27 | 67.88 | 0.4122 | |
| RNANTraining noise protocol=Gaussian noise, σ = 152023.03 | 26.26 | 67.43 | 0.429 | |
| SwinIRTraining noise protocol=Gaussian noise, σ = 152023.03 | 26.14 | 66.51 | 0.4485 | |
| DropoutTraining noise protocol=Gaussian noise, σ = 152023.03 | 26.1 | 66.96 | 0.4443 | |
| baselineTraining noise protocol=Gaussian noise, σ = 152023.03 | 25.91 | 65.35 | 0.4564 | |
| RestormerTraining noise protocol=Gaussian noise, σ = 152023.03 | 24.59 | 58.8 | 0.5375 |