ESRGAN: Enhanced Super-Resolution Generative Adversarial Networks

Benchmark Model Rank Results
image-super-resolution-on-bsd100-4x-upscalingSRGAN + Residual-in-Residual Dense Block#12PSNR: 27.85SSIM: 0.7455
image-super-resolution-on-ffhq-1024-x-1024-4xESRGAN#9FID: 72.73MS-SSIM: 0.782PSNR: 19.84SSIM: 0.353
image-super-resolution-on-ffhq-256-x-256-4xESRGAN#9FID: 166.36MS-SSIM: 0.747PSNR: 15.43SSIM: 0.267
image-super-resolution-on-ffhq-512-x-512-4xESRGAN#5PSNR: 27.134SSIM: 0.741MS-SSIM: 0.935LLE: 2.261FED: 0.1107
image-super-resolution-on-manga109-4xSRGAN + Residual-in-Residual Dense Block#25SSIM: 0.9196PSNR: 31.66
image-super-resolution-on-manga109-4xbicubic#43SSIM: 0.7866PSNR: 24.89
image-super-resolution-on-set14-4x-upscalingSRGAN + Residual-in-Residual Dense Block#26PSNR: 28.99SSIM: 0.7917
image-super-resolution-on-urban100-4xSRGAN + Residual-in-Residual Dense Block#20PSNR: 27.03SSIM: 0.8153
image-super-resolution-on-urban100-4xbicubic#57PSNR: 23.14SSIM: 0.6577
video-super-resolution-on-msu-video-upscalersESRGAN#25SSIM: 0.936PSNR: 27.29VMAF: 56.69
video-super-resolution-on-msu-vsr-benchmarkESRGAN#14Subjective score: 5.353ERQAv1.0: 0.7351 - LPIPS: 0.948SSIM: 0.808