FFDNet: Toward a Fast and Flexible Solution for CNN based Image Denoising

Benchmark Model Rank Results
color-image-denoising-on-cbsd68-sigma15FFDNet#8PSNR: 33.87
color-image-denoising-on-cbsd68-sigma25FFDNet#7PSNR: 31.21
color-image-denoising-on-cbsd68-sigma35FFDNet#3PSNR: 29.58
color-image-denoising-on-cbsd68-sigma50FFDNet#12PSNR: 27.96
color-image-denoising-on-cbsd68-sigma75FFDNet#3PSNR: 26.24
color-image-denoising-on-mcmaster-sigma15FFDNet#3PSNR: 34.66
color-image-denoising-on-mcmaster-sigma25FFDNet#3PSNR: 32.35
color-image-denoising-on-mcmaster-sigma50FFDNet#4PSNR: 29.18
color-image-denoising-on-urban100-sigma15-1FFDNet#6Average PSNR: 33.83
denoising-on-darmstadt-noise-datasetFFDNet#5PSNR: 34.40
grayscale-image-denoising-on-bsd68-sigma15FFDNet#12PSNR: 31.63
grayscale-image-denoising-on-bsd68-sigma25FFDNet#11PSNR: 29.19
grayscale-image-denoising-on-bsd68-sigma50FFDNet#12PSNR: 26.29
grayscale-image-denoising-on-set12-sigma15FFDNet#6PSNR: 25.49