Beyond a Gaussian Denoiser: Residual Learning of Deep CNN for Image Denoising

Open paper
Benchmark Model Rank Results
jpeg-artifact-correction-on-classic5-quality-40-grayscaleDnCNN-3#4PSNR: 33.77
jpeg-artifact-correction-on-live1-quality-40-grayscaleDnCNN-3#4PSNR: 33.96
jpeg-artifact-correction-on-classic5-quality-20-grayscaleDnCNN-3#5PSNR: 31.63
jpeg-artifact-correction-on-classic5-quality-30-grayscaleDnCNN-3#5PSNR: 32.91
jpeg-artifact-correction-on-live1-quality-30-grayscaleDnCNN-3#5PSNR: 32.98
color-image-denoising-on-cbsd68-sigma35DnCNN-B*#6PSNR: 28.74
denoising-on-darmstadt-noise-datasetCDnCNN-B#6PSNR: 32.43
grayscale-image-denoising-on-urban100-sigma15DnCNN#6PSNR: 32.67
jpeg-artifact-correction-on-classic5-quality-10-grayscaleDnCNN-3#6PSNR: 29.4
color-image-denoising-on-urban100-sigma15DnCNN#7Average PSNR: 32.98
jpeg-artifact-correction-on-live1-quality-10-grayscaleDnCNN-3#7PSNR: 29.19
jpeg-artifact-correction-on-live1-quality-20-grayscaleDnCNN-3#7PSNR: 31.59
grayscale-image-denoising-on-urban100-sigma25DnCNN#9PSNR: 29.97
grayscale-image-denoising-on-bsd68-sigma25DnCNN#10PSNR: 29.23
image-super-resolution-on-bsd100-3x-upscalingDnCNN-3#24PSNR: 28.85
image-super-resolution-on-set14-3x-upscalingDnCNN-3#25PSNR: 29.81
image-super-resolution-on-urban100-3x-upscalingDnCNN-3#25PSNR: 27.15
image-super-resolution-on-set5-3x-upscalingDnCNN-3#30PSNR: 33.75
image-super-resolution-on-bsd100-2x-upscalingDnCNN-3#34PSNR: 31.9
image-super-resolution-on-urban100-2x-upscalingDnCNN-3#35PSNR: 30.74
image-super-resolution-on-set5-2x-upscalingDnCNN-3#38PSNR: 37.58
image-super-resolution-on-set14-2x-upscalingDnCNN-3#39PSNR: 33.03
image-super-resolution-on-bsd100-4x-upscalingDnCNN-3#52PSNR: 27.29SSIM: 0.7253
image-super-resolution-on-urban100-4x-upscalingDnCNN-3#60PSNR: 25.2SSIM: 0.7521
image-super-resolution-on-set14-4x-upscalingDnCNN-3#75PSNR: 28.04SSIM: 0.7672

Rank counts only results with a code link.