Tasks SotA Datasets Papers Methods Submit About

Papers With Code 2

A community resource for machine learning research: papers, code, benchmarks, and state-of-the-art results.

Explore

Notable Benchmarks All SotA Datasets Papers Methods

Community

Submit Results About

Data sourced from the PWC Archive (CC-BY-SA 4.0). Built by the community, for the community.

Models/EDSR

EDSR

Reported on 118 benchmarks across 6 tasks · 1 paper · 62 SOTA

Note: results are matched by exact model name. Different papers may use the same name for different model variants.

Computer Vision59 results

Image RestorationonARAD-1K
MRAE· 2017-07-10
0.3277
best: 0.3814 (HSCNN+)
SOTA
Enhanced Deep Residual Networks for Single Image Super-Resolution arXiv:1707.02921
Image RestorationonARAD-1K
PSNR· 2017-07-10
28.29
best: 34.32 (MST++)
SOTA
Enhanced Deep Residual Networks for Single Image Super-Resolution arXiv:1707.02921
Image RestorationonARAD-1K
RMSE· 2017-07-10
0.0437
best: 0.0248 (MST++)
SOTA
Enhanced Deep Residual Networks for Single Image Super-Resolution arXiv:1707.02921
Image Super-ResolutiononSet5 - 4x upscaling
PSNR· 2017-07-10
32.46
best: 33.38 (HMA†)
SOTA
Enhanced Deep Residual Networks for Single Image Super-Resolution arXiv:1707.02921
Image Super-ResolutiononSet5 - 4x upscaling
SSIM· 2017-07-10
0.8968
best: 0.9103 (Hi-IR-L)
SOTA
Enhanced Deep Residual Networks for Single Image Super-Resolution arXiv:1707.02921
Image Super-ResolutiononFFHQ 512 x 512 - 4x upscaling
MS-SSIM· 2017-07-10
0.961
best: 0.976 (UNET++)
SOTA
Enhanced Deep Residual Networks for Single Image Super-Resolution arXiv:1707.02921
Image Super-ResolutiononFFHQ 512 x 512 - 4x upscaling
NIQE· 2017-07-10
13.636
best: 13.901 (SRFBN)
SOTA
Enhanced Deep Residual Networks for Single Image Super-Resolution arXiv:1707.02921
Image Super-ResolutiononFFHQ 512 x 512 - 4x upscaling
PSNR· 2017-07-10
30.188
best: 30.824 (HiFaceGAN)
SOTA
Enhanced Deep Residual Networks for Single Image Super-Resolution arXiv:1707.02921
Image Super-ResolutiononFFHQ 512 x 512 - 4x upscaling
SSIM· 2017-07-10
0.824
best: 0.859 (UNET++)
SOTA
Enhanced Deep Residual Networks for Single Image Super-Resolution arXiv:1707.02921
Image Super-ResolutiononDIV2K val - 4x upscaling
PSNR· 2017-07-10
29.25
SOTA
Enhanced Deep Residual Networks for Single Image Super-Resolution arXiv:1707.02921
Image Super-ResolutiononDIV2K val - 4x upscaling
SSIM· 2017-07-10
0.9017
SOTA
Enhanced Deep Residual Networks for Single Image Super-Resolution arXiv:1707.02921
Image Super-ResolutiononFFHQ 1024 x 1024 - 4x upscaling
FID· 2017-07-10
15.54
best: 1.978 (HiFaceGAN)
SOTA
Enhanced Deep Residual Networks for Single Image Super-Resolution arXiv:1707.02921
Image Super-ResolutiononManga109 - 4x upscaling
PSNR· 2017-07-10
31.02
best: 33.19 (HMA†)
SOTA
Enhanced Deep Residual Networks for Single Image Super-Resolution arXiv:1707.02921
Image Super-ResolutiononManga109 - 4x upscaling
SSIM· 2017-07-10
0.9148
best: 0.9366 (Hi-IR-L)
SOTA
Enhanced Deep Residual Networks for Single Image Super-Resolution arXiv:1707.02921
Image Super-ResolutiononUrban100 - 4x upscaling
PSNR· 2017-07-10
26.64
best: 28.72 (Hi-IR-L)
SOTA
Enhanced Deep Residual Networks for Single Image Super-Resolution arXiv:1707.02921
Image Super-ResolutiononUrban100 - 4x upscaling
SSIM· 2017-07-10
0.8033
best: 0.9481 (SPSR)
SOTA
Enhanced Deep Residual Networks for Single Image Super-Resolution arXiv:1707.02921
Image Super-ResolutiononBSD100 - 4x upscaling
PSNR· 2017-07-10
27.71
best: 28.16 (DRCT-L)
SOTA
Enhanced Deep Residual Networks for Single Image Super-Resolution arXiv:1707.02921
3D Object Super-ResolutiononSet5 - 4x upscaling
PSNR· 2017-07-10
32.46
best: 33.38 (HMA†)
SOTA
Enhanced Deep Residual Networks for Single Image Super-Resolution arXiv:1707.02921
3D Object Super-ResolutiononSet5 - 4x upscaling
SSIM· 2017-07-10
0.8968
best: 0.9103 (Hi-IR-L)
SOTA
Enhanced Deep Residual Networks for Single Image Super-Resolution arXiv:1707.02921
3D Object Super-ResolutiononFFHQ 512 x 512 - 4x upscaling
MS-SSIM· 2017-07-10
0.961
best: 0.976 (UNET++)
SOTA
Enhanced Deep Residual Networks for Single Image Super-Resolution arXiv:1707.02921
3D Object Super-ResolutiononFFHQ 512 x 512 - 4x upscaling
NIQE· 2017-07-10
13.636
best: 13.901 (SRFBN)
SOTA
Enhanced Deep Residual Networks for Single Image Super-Resolution arXiv:1707.02921
3D Object Super-ResolutiononFFHQ 512 x 512 - 4x upscaling
PSNR· 2017-07-10
30.188
best: 30.824 (HiFaceGAN)
SOTA
Enhanced Deep Residual Networks for Single Image Super-Resolution arXiv:1707.02921
3D Object Super-ResolutiononFFHQ 512 x 512 - 4x upscaling
SSIM· 2017-07-10
0.824
best: 0.859 (UNET++)
SOTA
Enhanced Deep Residual Networks for Single Image Super-Resolution arXiv:1707.02921
3D Object Super-ResolutiononDIV2K val - 4x upscaling
PSNR· 2017-07-10
29.25
SOTA
Enhanced Deep Residual Networks for Single Image Super-Resolution arXiv:1707.02921
3D Object Super-ResolutiononDIV2K val - 4x upscaling
SSIM· 2017-07-10
0.9017
SOTA
Enhanced Deep Residual Networks for Single Image Super-Resolution arXiv:1707.02921
3D Object Super-ResolutiononFFHQ 1024 x 1024 - 4x upscaling
FID· 2017-07-10
15.54
best: 1.978 (HiFaceGAN)
SOTA
Enhanced Deep Residual Networks for Single Image Super-Resolution arXiv:1707.02921
3D Object Super-ResolutiononManga109 - 4x upscaling
PSNR· 2017-07-10
31.02
best: 33.19 (HMA†)
SOTA
Enhanced Deep Residual Networks for Single Image Super-Resolution arXiv:1707.02921
3D Object Super-ResolutiononManga109 - 4x upscaling
SSIM· 2017-07-10
0.9148
best: 0.9366 (Hi-IR-L)
SOTA
Enhanced Deep Residual Networks for Single Image Super-Resolution arXiv:1707.02921
3D Object Super-ResolutiononUrban100 - 4x upscaling
PSNR· 2017-07-10
26.64
best: 28.72 (Hi-IR-L)
SOTA
Enhanced Deep Residual Networks for Single Image Super-Resolution arXiv:1707.02921
3D Object Super-ResolutiononUrban100 - 4x upscaling
SSIM· 2017-07-10
0.8033
best: 0.9481 (SPSR)
SOTA
Enhanced Deep Residual Networks for Single Image Super-Resolution arXiv:1707.02921
3D Object Super-ResolutiononBSD100 - 4x upscaling
PSNR· 2017-07-10
27.71
best: 28.16 (DRCT-L)
SOTA
Enhanced Deep Residual Networks for Single Image Super-Resolution arXiv:1707.02921
Image Super-ResolutiononSet14 - 4x upscaling
PSNR· 2017-07-10
28.8
best: 29.54 (DRCT-L)
Enhanced Deep Residual Networks for Single Image Super-Resolution arXiv:1707.02921
Image Super-ResolutiononSet14 - 4x upscaling
SSIM· 2017-07-10
0.7876
best: 0.894 (Edge-informed SR)
Enhanced Deep Residual Networks for Single Image Super-Resolution arXiv:1707.02921
Image Super-ResolutiononFFHQ 256 x 256 - 4x upscaling
FID· 2017-07-10
129.14
best: 5.36 (HiFaceGAN)
Enhanced Deep Residual Networks for Single Image Super-Resolution arXiv:1707.02921
Image Super-ResolutiononFFHQ 256 x 256 - 4x upscaling
MS-SSIM· 2017-07-10
0.901
best: 0.971 (HiFaceGAN)
Enhanced Deep Residual Networks for Single Image Super-Resolution arXiv:1707.02921
Image Super-ResolutiononFFHQ 256 x 256 - 4x upscaling
PSNR· 2017-07-10
22.47
best: 28.65 (HiFaceGAN)
Enhanced Deep Residual Networks for Single Image Super-Resolution arXiv:1707.02921
Image Super-ResolutiononFFHQ 256 x 256 - 4x upscaling
SSIM· 2017-07-10
0.706
best: 0.816 (HiFaceGAN)
Enhanced Deep Residual Networks for Single Image Super-Resolution arXiv:1707.02921
Image Super-ResolutiononFFHQ 512 x 512 - 4x upscaling
FED· 2017-07-10
0.0843
best: 0.1416 (Super-FAN)
Enhanced Deep Residual Networks for Single Image Super-Resolution arXiv:1707.02921
Image Super-ResolutiononFFHQ 512 x 512 - 4x upscaling
FID· 2017-07-10
20.605
best: 1.898 (HiFaceGAN)
Enhanced Deep Residual Networks for Single Image Super-Resolution arXiv:1707.02921
Image Super-ResolutiononFFHQ 512 x 512 - 4x upscaling
LLE· 2017-07-10
2.003
best: 2.702 (WaveletCNN)
Enhanced Deep Residual Networks for Single Image Super-Resolution arXiv:1707.02921
Image Super-ResolutiononFFHQ 512 x 512 - 4x upscaling
LPIPS· 2017-07-10
0.2475
best: 0.0723 (HiFaceGAN)
Enhanced Deep Residual Networks for Single Image Super-Resolution arXiv:1707.02921
Image Super-ResolutiononFFHQ 1024 x 1024 - 4x upscaling
MS-SSIM· 2017-07-10
0.933
best: 0.975 (HiFaceGAN)
Enhanced Deep Residual Networks for Single Image Super-Resolution arXiv:1707.02921
Image Super-ResolutiononFFHQ 1024 x 1024 - 4x upscaling
PSNR· 2017-07-10
28.34
best: 34.1 (CAGFace)
Enhanced Deep Residual Networks for Single Image Super-Resolution arXiv:1707.02921
Image Super-ResolutiononFFHQ 1024 x 1024 - 4x upscaling
SSIM· 2017-07-10
0.827
best: 0.906 (CAGFace)
Enhanced Deep Residual Networks for Single Image Super-Resolution arXiv:1707.02921
Image Super-ResolutiononBSD100 - 4x upscaling
SSIM· 2017-07-10
0.742
best: 0.851 (Edge-informed SR)
Enhanced Deep Residual Networks for Single Image Super-Resolution arXiv:1707.02921
3D Object Super-ResolutiononSet14 - 4x upscaling
PSNR· 2017-07-10
28.8
best: 29.54 (DRCT-L)
Enhanced Deep Residual Networks for Single Image Super-Resolution arXiv:1707.02921
3D Object Super-ResolutiononSet14 - 4x upscaling
SSIM· 2017-07-10
0.7876
best: 0.894 (Edge-informed SR)
Enhanced Deep Residual Networks for Single Image Super-Resolution arXiv:1707.02921
3D Object Super-ResolutiononFFHQ 256 x 256 - 4x upscaling
FID· 2017-07-10
129.14
best: 5.36 (HiFaceGAN)
Enhanced Deep Residual Networks for Single Image Super-Resolution arXiv:1707.02921
3D Object Super-ResolutiononFFHQ 256 x 256 - 4x upscaling
MS-SSIM· 2017-07-10
0.901
best: 0.971 (HiFaceGAN)
Enhanced Deep Residual Networks for Single Image Super-Resolution arXiv:1707.02921
3D Object Super-ResolutiononFFHQ 256 x 256 - 4x upscaling
PSNR· 2017-07-10
22.47
best: 28.65 (HiFaceGAN)
Enhanced Deep Residual Networks for Single Image Super-Resolution arXiv:1707.02921
3D Object Super-ResolutiononFFHQ 256 x 256 - 4x upscaling
SSIM· 2017-07-10
0.706
best: 0.816 (HiFaceGAN)
Enhanced Deep Residual Networks for Single Image Super-Resolution arXiv:1707.02921
3D Object Super-ResolutiononFFHQ 512 x 512 - 4x upscaling
FED· 2017-07-10
0.0843
best: 0.1416 (Super-FAN)
Enhanced Deep Residual Networks for Single Image Super-Resolution arXiv:1707.02921
3D Object Super-ResolutiononFFHQ 512 x 512 - 4x upscaling
FID· 2017-07-10
20.605
best: 1.898 (HiFaceGAN)
Enhanced Deep Residual Networks for Single Image Super-Resolution arXiv:1707.02921
3D Object Super-ResolutiononFFHQ 512 x 512 - 4x upscaling
LLE· 2017-07-10
2.003
best: 2.702 (WaveletCNN)
Enhanced Deep Residual Networks for Single Image Super-Resolution arXiv:1707.02921
3D Object Super-ResolutiononFFHQ 512 x 512 - 4x upscaling
LPIPS· 2017-07-10
0.2475
best: 0.0723 (HiFaceGAN)
Enhanced Deep Residual Networks for Single Image Super-Resolution arXiv:1707.02921
3D Object Super-ResolutiononFFHQ 1024 x 1024 - 4x upscaling
MS-SSIM· 2017-07-10
0.933
best: 0.975 (HiFaceGAN)
Enhanced Deep Residual Networks for Single Image Super-Resolution arXiv:1707.02921
3D Object Super-ResolutiononFFHQ 1024 x 1024 - 4x upscaling
PSNR· 2017-07-10
28.34
best: 34.1 (CAGFace)
Enhanced Deep Residual Networks for Single Image Super-Resolution arXiv:1707.02921
3D Object Super-ResolutiononFFHQ 1024 x 1024 - 4x upscaling
SSIM· 2017-07-10
0.827
best: 0.906 (CAGFace)
Enhanced Deep Residual Networks for Single Image Super-Resolution arXiv:1707.02921
3D Object Super-ResolutiononBSD100 - 4x upscaling
SSIM· 2017-07-10
0.742
best: 0.851 (Edge-informed SR)
Enhanced Deep Residual Networks for Single Image Super-Resolution arXiv:1707.02921

Graphs28 results

Super-ResolutiononSet5 - 4x upscaling
PSNR· 2017-07-10
32.46
best: 33.38 (HMA†)
SOTA
Enhanced Deep Residual Networks for Single Image Super-Resolution arXiv:1707.02921
Super-ResolutiononSet5 - 4x upscaling
SSIM· 2017-07-10
0.8968
best: 0.9103 (Hi-IR-L)
SOTA
Enhanced Deep Residual Networks for Single Image Super-Resolution arXiv:1707.02921
Super-ResolutiononFFHQ 512 x 512 - 4x upscaling
MS-SSIM· 2017-07-10
0.961
best: 0.976 (UNET++)
SOTA
Enhanced Deep Residual Networks for Single Image Super-Resolution arXiv:1707.02921
Super-ResolutiononFFHQ 512 x 512 - 4x upscaling
NIQE· 2017-07-10
13.636
best: 13.901 (SRFBN)
SOTA
Enhanced Deep Residual Networks for Single Image Super-Resolution arXiv:1707.02921
Super-ResolutiononFFHQ 512 x 512 - 4x upscaling
PSNR· 2017-07-10
30.188
best: 30.824 (HiFaceGAN)
SOTA
Enhanced Deep Residual Networks for Single Image Super-Resolution arXiv:1707.02921
Super-ResolutiononFFHQ 512 x 512 - 4x upscaling
SSIM· 2017-07-10
0.824
best: 0.859 (UNET++)
SOTA
Enhanced Deep Residual Networks for Single Image Super-Resolution arXiv:1707.02921
Super-ResolutiononDIV2K val - 4x upscaling
PSNR· 2017-07-10
29.25
SOTA
Enhanced Deep Residual Networks for Single Image Super-Resolution arXiv:1707.02921
Super-ResolutiononDIV2K val - 4x upscaling
SSIM· 2017-07-10
0.9017
SOTA
Enhanced Deep Residual Networks for Single Image Super-Resolution arXiv:1707.02921
Super-ResolutiononFFHQ 1024 x 1024 - 4x upscaling
FID· 2017-07-10
15.54
best: 1.978 (HiFaceGAN)
SOTA
Enhanced Deep Residual Networks for Single Image Super-Resolution arXiv:1707.02921
Super-ResolutiononManga109 - 4x upscaling
PSNR· 2017-07-10
31.02
best: 33.19 (HMA†)
SOTA
Enhanced Deep Residual Networks for Single Image Super-Resolution arXiv:1707.02921
Super-ResolutiononManga109 - 4x upscaling
SSIM· 2017-07-10
0.9148
best: 0.9366 (Hi-IR-L)
SOTA
Enhanced Deep Residual Networks for Single Image Super-Resolution arXiv:1707.02921
Super-ResolutiononUrban100 - 4x upscaling
PSNR· 2017-07-10
26.64
best: 28.72 (Hi-IR-L)
SOTA
Enhanced Deep Residual Networks for Single Image Super-Resolution arXiv:1707.02921
Super-ResolutiononUrban100 - 4x upscaling
SSIM· 2017-07-10
0.8033
best: 0.9481 (SPSR)
SOTA
Enhanced Deep Residual Networks for Single Image Super-Resolution arXiv:1707.02921
Super-ResolutiononBSD100 - 4x upscaling
PSNR· 2017-07-10
27.71
best: 28.16 (DRCT-L)
SOTA
Enhanced Deep Residual Networks for Single Image Super-Resolution arXiv:1707.02921
Super-ResolutiononSet14 - 4x upscaling
PSNR· 2017-07-10
28.8
best: 29.54 (DRCT-L)
Enhanced Deep Residual Networks for Single Image Super-Resolution arXiv:1707.02921
Super-ResolutiononSet14 - 4x upscaling
SSIM· 2017-07-10
0.7876
best: 0.894 (Edge-informed SR)
Enhanced Deep Residual Networks for Single Image Super-Resolution arXiv:1707.02921
Super-ResolutiononFFHQ 256 x 256 - 4x upscaling
FID· 2017-07-10
129.14
best: 5.36 (HiFaceGAN)
Enhanced Deep Residual Networks for Single Image Super-Resolution arXiv:1707.02921
Super-ResolutiononFFHQ 256 x 256 - 4x upscaling
MS-SSIM· 2017-07-10
0.901
best: 0.971 (HiFaceGAN)
Enhanced Deep Residual Networks for Single Image Super-Resolution arXiv:1707.02921
Super-ResolutiononFFHQ 256 x 256 - 4x upscaling
PSNR· 2017-07-10
22.47
best: 28.65 (HiFaceGAN)
Enhanced Deep Residual Networks for Single Image Super-Resolution arXiv:1707.02921
Super-ResolutiononFFHQ 256 x 256 - 4x upscaling
SSIM· 2017-07-10
0.706
best: 0.816 (HiFaceGAN)
Enhanced Deep Residual Networks for Single Image Super-Resolution arXiv:1707.02921
Super-ResolutiononFFHQ 512 x 512 - 4x upscaling
FED· 2017-07-10
0.0843
best: 0.1416 (Super-FAN)
Enhanced Deep Residual Networks for Single Image Super-Resolution arXiv:1707.02921
Super-ResolutiononFFHQ 512 x 512 - 4x upscaling
FID· 2017-07-10
20.605
best: 1.898 (HiFaceGAN)
Enhanced Deep Residual Networks for Single Image Super-Resolution arXiv:1707.02921
Super-ResolutiononFFHQ 512 x 512 - 4x upscaling
LLE· 2017-07-10
2.003
best: 2.702 (WaveletCNN)
Enhanced Deep Residual Networks for Single Image Super-Resolution arXiv:1707.02921
Super-ResolutiononFFHQ 512 x 512 - 4x upscaling
LPIPS· 2017-07-10
0.2475
best: 0.0723 (HiFaceGAN)
Enhanced Deep Residual Networks for Single Image Super-Resolution arXiv:1707.02921
Super-ResolutiononFFHQ 1024 x 1024 - 4x upscaling
MS-SSIM· 2017-07-10
0.933
best: 0.975 (HiFaceGAN)
Enhanced Deep Residual Networks for Single Image Super-Resolution arXiv:1707.02921
Super-ResolutiononFFHQ 1024 x 1024 - 4x upscaling
PSNR· 2017-07-10
28.34
best: 34.1 (CAGFace)
Enhanced Deep Residual Networks for Single Image Super-Resolution arXiv:1707.02921
Super-ResolutiononFFHQ 1024 x 1024 - 4x upscaling
SSIM· 2017-07-10
0.827
best: 0.906 (CAGFace)
Enhanced Deep Residual Networks for Single Image Super-Resolution arXiv:1707.02921
Super-ResolutiononBSD100 - 4x upscaling
SSIM· 2017-07-10
0.742
best: 0.851 (Edge-informed SR)
Enhanced Deep Residual Networks for Single Image Super-Resolution arXiv:1707.02921

Methodology28 results

16konSet5 - 4x upscaling
PSNR· 2017-07-10
32.46
best: 33.38 (HMA†)
SOTA
Enhanced Deep Residual Networks for Single Image Super-Resolution arXiv:1707.02921
16konSet5 - 4x upscaling
SSIM· 2017-07-10
0.8968
best: 0.9103 (Hi-IR-L)
SOTA
Enhanced Deep Residual Networks for Single Image Super-Resolution arXiv:1707.02921
16konFFHQ 512 x 512 - 4x upscaling
MS-SSIM· 2017-07-10
0.961
best: 0.976 (UNET++)
SOTA
Enhanced Deep Residual Networks for Single Image Super-Resolution arXiv:1707.02921
16konFFHQ 512 x 512 - 4x upscaling
NIQE· 2017-07-10
13.636
best: 13.901 (SRFBN)
SOTA
Enhanced Deep Residual Networks for Single Image Super-Resolution arXiv:1707.02921
16konFFHQ 512 x 512 - 4x upscaling
PSNR· 2017-07-10
30.188
best: 30.824 (HiFaceGAN)
SOTA
Enhanced Deep Residual Networks for Single Image Super-Resolution arXiv:1707.02921
16konFFHQ 512 x 512 - 4x upscaling
SSIM· 2017-07-10
0.824
best: 0.859 (UNET++)
SOTA
Enhanced Deep Residual Networks for Single Image Super-Resolution arXiv:1707.02921
16konDIV2K val - 4x upscaling
PSNR· 2017-07-10
29.25
SOTA
Enhanced Deep Residual Networks for Single Image Super-Resolution arXiv:1707.02921
16konDIV2K val - 4x upscaling
SSIM· 2017-07-10
0.9017
SOTA
Enhanced Deep Residual Networks for Single Image Super-Resolution arXiv:1707.02921
16konFFHQ 1024 x 1024 - 4x upscaling
FID· 2017-07-10
15.54
best: 1.978 (HiFaceGAN)
SOTA
Enhanced Deep Residual Networks for Single Image Super-Resolution arXiv:1707.02921
16konManga109 - 4x upscaling
PSNR· 2017-07-10
31.02
best: 33.19 (HMA†)
SOTA
Enhanced Deep Residual Networks for Single Image Super-Resolution arXiv:1707.02921
16konManga109 - 4x upscaling
SSIM· 2017-07-10
0.9148
best: 0.9366 (Hi-IR-L)
SOTA
Enhanced Deep Residual Networks for Single Image Super-Resolution arXiv:1707.02921
16konUrban100 - 4x upscaling
PSNR· 2017-07-10
26.64
best: 28.72 (Hi-IR-L)
SOTA
Enhanced Deep Residual Networks for Single Image Super-Resolution arXiv:1707.02921
16konUrban100 - 4x upscaling
SSIM· 2017-07-10
0.8033
best: 0.9481 (SPSR)
SOTA
Enhanced Deep Residual Networks for Single Image Super-Resolution arXiv:1707.02921
16konBSD100 - 4x upscaling
PSNR· 2017-07-10
27.71
best: 28.16 (DRCT-L)
SOTA
Enhanced Deep Residual Networks for Single Image Super-Resolution arXiv:1707.02921
16konSet14 - 4x upscaling
PSNR· 2017-07-10
28.8
best: 29.54 (DRCT-L)
Enhanced Deep Residual Networks for Single Image Super-Resolution arXiv:1707.02921
16konSet14 - 4x upscaling
SSIM· 2017-07-10
0.7876
best: 0.894 (Edge-informed SR)
Enhanced Deep Residual Networks for Single Image Super-Resolution arXiv:1707.02921
16konFFHQ 256 x 256 - 4x upscaling
FID· 2017-07-10
129.14
best: 5.36 (HiFaceGAN)
Enhanced Deep Residual Networks for Single Image Super-Resolution arXiv:1707.02921
16konFFHQ 256 x 256 - 4x upscaling
MS-SSIM· 2017-07-10
0.901
best: 0.971 (HiFaceGAN)
Enhanced Deep Residual Networks for Single Image Super-Resolution arXiv:1707.02921
16konFFHQ 256 x 256 - 4x upscaling
PSNR· 2017-07-10
22.47
best: 28.65 (HiFaceGAN)
Enhanced Deep Residual Networks for Single Image Super-Resolution arXiv:1707.02921
16konFFHQ 256 x 256 - 4x upscaling
SSIM· 2017-07-10
0.706
best: 0.816 (HiFaceGAN)
Enhanced Deep Residual Networks for Single Image Super-Resolution arXiv:1707.02921
16konFFHQ 512 x 512 - 4x upscaling
FED· 2017-07-10
0.0843
best: 0.1416 (Super-FAN)
Enhanced Deep Residual Networks for Single Image Super-Resolution arXiv:1707.02921
16konFFHQ 512 x 512 - 4x upscaling
FID· 2017-07-10
20.605
best: 1.898 (HiFaceGAN)
Enhanced Deep Residual Networks for Single Image Super-Resolution arXiv:1707.02921
16konFFHQ 512 x 512 - 4x upscaling
LLE· 2017-07-10
2.003
best: 2.702 (WaveletCNN)
Enhanced Deep Residual Networks for Single Image Super-Resolution arXiv:1707.02921
16konFFHQ 512 x 512 - 4x upscaling
LPIPS· 2017-07-10
0.2475
best: 0.0723 (HiFaceGAN)
Enhanced Deep Residual Networks for Single Image Super-Resolution arXiv:1707.02921
16konFFHQ 1024 x 1024 - 4x upscaling
MS-SSIM· 2017-07-10
0.933
best: 0.975 (HiFaceGAN)
Enhanced Deep Residual Networks for Single Image Super-Resolution arXiv:1707.02921
16konFFHQ 1024 x 1024 - 4x upscaling
PSNR· 2017-07-10
28.34
best: 34.1 (CAGFace)
Enhanced Deep Residual Networks for Single Image Super-Resolution arXiv:1707.02921
16konFFHQ 1024 x 1024 - 4x upscaling
SSIM· 2017-07-10
0.827
best: 0.906 (CAGFace)
Enhanced Deep Residual Networks for Single Image Super-Resolution arXiv:1707.02921
16konBSD100 - 4x upscaling
SSIM· 2017-07-10
0.742
best: 0.851 (Edge-informed SR)
Enhanced Deep Residual Networks for Single Image Super-Resolution arXiv:1707.02921

Audio3 results

10-shot image generationonARAD-1K
MRAE· 2017-07-10
0.3277
best: 0.3814 (HSCNN+)
SOTA
Enhanced Deep Residual Networks for Single Image Super-Resolution arXiv:1707.02921
10-shot image generationonARAD-1K
PSNR· 2017-07-10
28.29
best: 34.32 (MST++)
SOTA
Enhanced Deep Residual Networks for Single Image Super-Resolution arXiv:1707.02921
10-shot image generationonARAD-1K
RMSE· 2017-07-10
0.0437
best: 0.0248 (MST++)
SOTA
Enhanced Deep Residual Networks for Single Image Super-Resolution arXiv:1707.02921