Image Super-Resolution Using Deep Convolutional Networks

Benchmark Model Rank Results
image-super-resolution-on-bsd100-4x-upscalingSRCNN#50PSNR: 26.9SSIM: 0.7101
image-super-resolution-on-ffhq-1024-x-1024-4xSRCNN#7FID: 31.84MS-SSIM: 0.924PSNR: 27.40SSIM: 0.801
image-super-resolution-on-ffhq-256-x-256-4xSRCNN#7FID: 147.21MS-SSIM: 0.900PSNR: 23.12SSIM: 0.688
image-super-resolution-on-ixiSRCNN#6PSNR 2x T2w: 37.32PSNR 4x T2w: 29.69SSIM 4x T2w: 0.9052
image-super-resolution-on-manga109-4xSRCNN#41SSIM: 0.8555PSNR: 27.58
image-super-resolution-on-set14-4x-upscalingSRCNN#80PSNR: 27.5SSIM: 0.7513
image-super-resolution-on-set5-4x-upscalingSRCNN#8PSNR: 30.49SSIM: 0.8628
image-super-resolution-on-urban100-4xSRCNN#54PSNR: 24.52SSIM: 0.7221
video-super-resolution-on-msu-video-upscalersSRCNN#20SSIM: 0.929PSNR: 26.68VMAF: 51.21
video-super-resolution-on-vid4-4x-upscalingSRCNN#20SSIM: 0.7158PSNR: 24.68MOVIE: 6.9