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/ESRGAN

ESRGAN

Reported on 201 benchmarks across 17 tasks · 2 papers · 45 SOTA

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

Computer Vision94 results

3D Human Pose EstimationonMSU Video Super Resolution Benchmark: Detail Restoration
ERQAv1.0· 2018-09-01
0.735
best: 0.758 (VRT)
SOTA
ESRGAN: Enhanced Super-Resolution Generative Adversarial Networks arXiv:1809.00219
3D Human Pose EstimationonMSU Video Super Resolution Benchmark: Detail Restoration
Subjective score· 2018-09-01
5.353
best: 7.628 (VRT)
SOTA
ESRGAN: Enhanced Super-Resolution Generative Adversarial Networks arXiv:1809.00219
VideoonMSU Video Super Resolution Benchmark: Detail Restoration
ERQAv1.0· 2018-09-01
0.735
best: 0.758 (VRT)
SOTA
ESRGAN: Enhanced Super-Resolution Generative Adversarial Networks arXiv:1809.00219
VideoonMSU Video Super Resolution Benchmark: Detail Restoration
Subjective score· 2018-09-01
5.353
best: 7.628 (VRT)
SOTA
ESRGAN: Enhanced Super-Resolution Generative Adversarial Networks arXiv:1809.00219
Pose EstimationonMSU Video Super Resolution Benchmark: Detail Restoration
ERQAv1.0· 2018-09-01
0.735
best: 0.758 (VRT)
SOTA
ESRGAN: Enhanced Super-Resolution Generative Adversarial Networks arXiv:1809.00219
Pose EstimationonMSU Video Super Resolution Benchmark: Detail Restoration
Subjective score· 2018-09-01
5.353
best: 7.628 (VRT)
SOTA
ESRGAN: Enhanced Super-Resolution Generative Adversarial Networks arXiv:1809.00219
Face ReconstructiononFFHQ 512 x 512 - 16x upscaling
FID· 2018-09-01
50.901
best: 11.389 (HiFaceGAN)
SOTA
ESRGAN: Enhanced Super-Resolution Generative Adversarial Networks arXiv:1809.00219
Face ReconstructiononFFHQ 512 x 512 - 16x upscaling
LPIPS· 2018-09-01
0.3928
best: 0.2449 (HiFaceGAN)
SOTA
ESRGAN: Enhanced Super-Resolution Generative Adversarial Networks arXiv:1809.00219
Face ReconstructiononFFHQ 512 x 512 - 16x upscaling
NIQE· 2018-09-01
15.383
SOTA
ESRGAN: Enhanced Super-Resolution Generative Adversarial Networks arXiv:1809.00219
Image Super-ResolutiononFFHQ 512 x 512 - 4x upscaling
FID· 2018-09-01
3.503
best: 1.898 (HiFaceGAN)
SOTA
ESRGAN: Enhanced Super-Resolution Generative Adversarial Networks arXiv:1809.00219
Image Super-ResolutiononFFHQ 512 x 512 - 4x upscaling
LPIPS· 2018-09-01
0.1221
best: 0.0723 (HiFaceGAN)
SOTA
ESRGAN: Enhanced Super-Resolution Generative Adversarial Networks arXiv:1809.00219
3D Face ReconstructiononFFHQ 512 x 512 - 16x upscaling
FID· 2018-09-01
50.901
best: 11.389 (HiFaceGAN)
SOTA
ESRGAN: Enhanced Super-Resolution Generative Adversarial Networks arXiv:1809.00219
3D Face ReconstructiononFFHQ 512 x 512 - 16x upscaling
LPIPS· 2018-09-01
0.3928
best: 0.2449 (HiFaceGAN)
SOTA
ESRGAN: Enhanced Super-Resolution Generative Adversarial Networks arXiv:1809.00219
3D Face ReconstructiononFFHQ 512 x 512 - 16x upscaling
NIQE· 2018-09-01
15.383
SOTA
ESRGAN: Enhanced Super-Resolution Generative Adversarial Networks arXiv:1809.00219
Video Super-ResolutiononMSU Video Super Resolution Benchmark: Detail Restoration
ERQAv1.0· 2018-09-01
0.735
best: 0.758 (VRT)
SOTA
ESRGAN: Enhanced Super-Resolution Generative Adversarial Networks arXiv:1809.00219
Video Super-ResolutiononMSU Video Super Resolution Benchmark: Detail Restoration
Subjective score· 2018-09-01
5.353
best: 7.628 (VRT)
SOTA
ESRGAN: Enhanced Super-Resolution Generative Adversarial Networks arXiv:1809.00219
3D Object Super-ResolutiononFFHQ 512 x 512 - 4x upscaling
FID· 2018-09-01
3.503
best: 1.898 (HiFaceGAN)
SOTA
ESRGAN: Enhanced Super-Resolution Generative Adversarial Networks arXiv:1809.00219
3D Object Super-ResolutiononFFHQ 512 x 512 - 4x upscaling
LPIPS· 2018-09-01
0.1221
best: 0.0723 (HiFaceGAN)
SOTA
ESRGAN: Enhanced Super-Resolution Generative Adversarial Networks arXiv:1809.00219
3D Object Super-ResolutiononMSU Video Super Resolution Benchmark: Detail Restoration
ERQAv1.0· 2018-09-01
0.735
best: 0.758 (VRT)
SOTA
ESRGAN: Enhanced Super-Resolution Generative Adversarial Networks arXiv:1809.00219
3D Object Super-ResolutiononMSU Video Super Resolution Benchmark: Detail Restoration
Subjective score· 2018-09-01
5.353
best: 7.628 (VRT)
SOTA
ESRGAN: Enhanced Super-Resolution Generative Adversarial Networks arXiv:1809.00219
Image Super-ResolutiononDIV2K val - 4x upscaling
PSNR· uses extra data· 2023-03-29
26.22
best: 29.25 (EDSR)
Implicit Diffusion Models for Continuous Super-Resolution arXiv:2303.16491
Image Super-ResolutiononDIV2K val - 4x upscaling
SSIM· uses extra data· 2023-03-29
0.75
best: 0.9017 (EDSR)
Implicit Diffusion Models for Continuous Super-Resolution arXiv:2303.16491
3D Object Super-ResolutiononDIV2K val - 4x upscaling
PSNR· uses extra data· 2023-03-29
26.22
best: 29.25 (EDSR)
Implicit Diffusion Models for Continuous Super-Resolution arXiv:2303.16491
3D Object Super-ResolutiononDIV2K val - 4x upscaling
SSIM· uses extra data· 2023-03-29
0.75
best: 0.9017 (EDSR)
Implicit Diffusion Models for Continuous Super-Resolution arXiv:2303.16491
3D Human Pose EstimationonMSU Video Upscalers: Quality Enhancement
PSNR· 2018-09-01
27.29
best: 31.28 (VEAI-AHQ-12)
ESRGAN: Enhanced Super-Resolution Generative Adversarial Networks arXiv:1809.00219
3D Human Pose EstimationonMSU Video Upscalers: Quality Enhancement
SSIM· 2018-09-01
0.936
best: 0.939 (iSeeBetter)
ESRGAN: Enhanced Super-Resolution Generative Adversarial Networks arXiv:1809.00219
3D Human Pose EstimationonMSU Video Upscalers: Quality Enhancement
VMAF· 2018-09-01
56.69
best: 61.2 (TecoGAN)
ESRGAN: Enhanced Super-Resolution Generative Adversarial Networks arXiv:1809.00219
3D Human Pose EstimationonMSU Video Super Resolution Benchmark: Detail Restoration
1 - LPIPS· 2018-09-01
0.948
best: 0.623 (DFDnet)
ESRGAN: Enhanced Super-Resolution Generative Adversarial Networks arXiv:1809.00219
3D Human Pose EstimationonMSU Video Super Resolution Benchmark: Detail Restoration
FPS· 2018-09-01
1.004
best: 5.882 (SRMD)
ESRGAN: Enhanced Super-Resolution Generative Adversarial Networks arXiv:1809.00219
3D Human Pose EstimationonMSU Video Super Resolution Benchmark: Detail Restoration
PSNR· 2018-09-01
27.33
best: 31.669 (VRT)
ESRGAN: Enhanced Super-Resolution Generative Adversarial Networks arXiv:1809.00219
3D Human Pose EstimationonMSU Video Super Resolution Benchmark: Detail Restoration
SSIM· 2018-09-01
0.808
best: 0.902 (VRT)
ESRGAN: Enhanced Super-Resolution Generative Adversarial Networks arXiv:1809.00219
VideoonMSU Video Upscalers: Quality Enhancement
PSNR· 2018-09-01
27.29
best: 31.28 (VEAI-AHQ-12)
ESRGAN: Enhanced Super-Resolution Generative Adversarial Networks arXiv:1809.00219
VideoonMSU Video Upscalers: Quality Enhancement
SSIM· 2018-09-01
0.936
best: 0.939 (iSeeBetter)
ESRGAN: Enhanced Super-Resolution Generative Adversarial Networks arXiv:1809.00219
VideoonMSU Video Upscalers: Quality Enhancement
VMAF· 2018-09-01
56.69
best: 61.2 (TecoGAN)
ESRGAN: Enhanced Super-Resolution Generative Adversarial Networks arXiv:1809.00219
VideoonMSU Video Super Resolution Benchmark: Detail Restoration
1 - LPIPS· 2018-09-01
0.948
best: 0.623 (DFDnet)
ESRGAN: Enhanced Super-Resolution Generative Adversarial Networks arXiv:1809.00219
VideoonMSU Video Super Resolution Benchmark: Detail Restoration
FPS· 2018-09-01
1.004
best: 5.882 (SRMD)
ESRGAN: Enhanced Super-Resolution Generative Adversarial Networks arXiv:1809.00219
VideoonMSU Video Super Resolution Benchmark: Detail Restoration
PSNR· 2018-09-01
27.33
best: 31.669 (VRT)
ESRGAN: Enhanced Super-Resolution Generative Adversarial Networks arXiv:1809.00219
VideoonMSU Video Super Resolution Benchmark: Detail Restoration
SSIM· 2018-09-01
0.808
best: 0.902 (VRT)
ESRGAN: Enhanced Super-Resolution Generative Adversarial Networks arXiv:1809.00219
Pose EstimationonMSU Video Upscalers: Quality Enhancement
PSNR· 2018-09-01
27.29
best: 31.28 (VEAI-AHQ-12)
ESRGAN: Enhanced Super-Resolution Generative Adversarial Networks arXiv:1809.00219
Pose EstimationonMSU Video Upscalers: Quality Enhancement
SSIM· 2018-09-01
0.936
best: 0.939 (iSeeBetter)
ESRGAN: Enhanced Super-Resolution Generative Adversarial Networks arXiv:1809.00219
Pose EstimationonMSU Video Upscalers: Quality Enhancement
VMAF· 2018-09-01
56.69
best: 61.2 (TecoGAN)
ESRGAN: Enhanced Super-Resolution Generative Adversarial Networks arXiv:1809.00219
Pose EstimationonMSU Video Super Resolution Benchmark: Detail Restoration
1 - LPIPS· 2018-09-01
0.948
best: 0.623 (DFDnet)
ESRGAN: Enhanced Super-Resolution Generative Adversarial Networks arXiv:1809.00219
Pose EstimationonMSU Video Super Resolution Benchmark: Detail Restoration
FPS· 2018-09-01
1.004
best: 5.882 (SRMD)
ESRGAN: Enhanced Super-Resolution Generative Adversarial Networks arXiv:1809.00219
Pose EstimationonMSU Video Super Resolution Benchmark: Detail Restoration
PSNR· 2018-09-01
27.33
best: 31.669 (VRT)
ESRGAN: Enhanced Super-Resolution Generative Adversarial Networks arXiv:1809.00219
Pose EstimationonMSU Video Super Resolution Benchmark: Detail Restoration
SSIM· 2018-09-01
0.808
best: 0.902 (VRT)
ESRGAN: Enhanced Super-Resolution Generative Adversarial Networks arXiv:1809.00219
Image Super-ResolutiononFFHQ 256 x 256 - 4x upscaling
FID· 2018-09-01
166.36
best: 5.36 (HiFaceGAN)
ESRGAN: Enhanced Super-Resolution Generative Adversarial Networks arXiv:1809.00219
Image Super-ResolutiononFFHQ 256 x 256 - 4x upscaling
MS-SSIM· 2018-09-01
0.747
best: 0.971 (HiFaceGAN)
ESRGAN: Enhanced Super-Resolution Generative Adversarial Networks arXiv:1809.00219
Image Super-ResolutiononFFHQ 256 x 256 - 4x upscaling
PSNR· 2018-09-01
15.43
best: 28.65 (HiFaceGAN)
ESRGAN: Enhanced Super-Resolution Generative Adversarial Networks arXiv:1809.00219
Image Super-ResolutiononFFHQ 256 x 256 - 4x upscaling
SSIM· 2018-09-01
0.267
best: 0.816 (HiFaceGAN)
ESRGAN: Enhanced Super-Resolution Generative Adversarial Networks arXiv:1809.00219
Image Super-ResolutiononFFHQ 512 x 512 - 4x upscaling
FED· 2018-09-01
0.1107
best: 0.1416 (Super-FAN)
ESRGAN: Enhanced Super-Resolution Generative Adversarial Networks arXiv:1809.00219
Image Super-ResolutiononFFHQ 512 x 512 - 4x upscaling
LLE· 2018-09-01
2.261
best: 2.702 (WaveletCNN)
ESRGAN: Enhanced Super-Resolution Generative Adversarial Networks arXiv:1809.00219
Image Super-ResolutiononFFHQ 512 x 512 - 4x upscaling
MS-SSIM· 2018-09-01
0.935
best: 0.976 (UNET++)
ESRGAN: Enhanced Super-Resolution Generative Adversarial Networks arXiv:1809.00219
Image Super-ResolutiononFFHQ 512 x 512 - 4x upscaling
NIQE· 2018-09-01
6.984
best: 13.901 (SRFBN)
ESRGAN: Enhanced Super-Resolution Generative Adversarial Networks arXiv:1809.00219
Image Super-ResolutiononFFHQ 512 x 512 - 4x upscaling
PSNR· 2018-09-01
27.134
best: 30.824 (HiFaceGAN)
ESRGAN: Enhanced Super-Resolution Generative Adversarial Networks arXiv:1809.00219
Image Super-ResolutiononFFHQ 512 x 512 - 4x upscaling
SSIM· 2018-09-01
0.741
best: 0.859 (UNET++)
ESRGAN: Enhanced Super-Resolution Generative Adversarial Networks arXiv:1809.00219
Image Super-ResolutiononFFHQ 1024 x 1024 - 4x upscaling
FID· 2018-09-01
72.73
best: 1.978 (HiFaceGAN)
ESRGAN: Enhanced Super-Resolution Generative Adversarial Networks arXiv:1809.00219
Image Super-ResolutiononFFHQ 1024 x 1024 - 4x upscaling
MS-SSIM· 2018-09-01
0.782
best: 0.975 (HiFaceGAN)
ESRGAN: Enhanced Super-Resolution Generative Adversarial Networks arXiv:1809.00219
Image Super-ResolutiononFFHQ 1024 x 1024 - 4x upscaling
PSNR· 2018-09-01
19.84
best: 34.1 (CAGFace)
ESRGAN: Enhanced Super-Resolution Generative Adversarial Networks arXiv:1809.00219
Image Super-ResolutiononFFHQ 1024 x 1024 - 4x upscaling
SSIM· 2018-09-01
0.353
best: 0.906 (CAGFace)
ESRGAN: Enhanced Super-Resolution Generative Adversarial Networks arXiv:1809.00219
Image Super-ResolutiononPIRM-test
NIQE· 2018-09-01
2.55
best: 2.71 (SRGAN)
ESRGAN: Enhanced Super-Resolution Generative Adversarial Networks arXiv:1809.00219
Video Super-ResolutiononMSU Video Upscalers: Quality Enhancement
PSNR· 2018-09-01
27.29
best: 31.28 (VEAI-AHQ-12)
ESRGAN: Enhanced Super-Resolution Generative Adversarial Networks arXiv:1809.00219
Video Super-ResolutiononMSU Video Upscalers: Quality Enhancement
SSIM· 2018-09-01
0.936
best: 0.939 (iSeeBetter)
ESRGAN: Enhanced Super-Resolution Generative Adversarial Networks arXiv:1809.00219
Video Super-ResolutiononMSU Video Upscalers: Quality Enhancement
VMAF· 2018-09-01
56.69
best: 61.2 (TecoGAN)
ESRGAN: Enhanced Super-Resolution Generative Adversarial Networks arXiv:1809.00219
Video Super-ResolutiononMSU Video Super Resolution Benchmark: Detail Restoration
1 - LPIPS· 2018-09-01
0.948
best: 0.623 (DFDnet)
ESRGAN: Enhanced Super-Resolution Generative Adversarial Networks arXiv:1809.00219
Video Super-ResolutiononMSU Video Super Resolution Benchmark: Detail Restoration
FPS· 2018-09-01
1.004
best: 5.882 (SRMD)
ESRGAN: Enhanced Super-Resolution Generative Adversarial Networks arXiv:1809.00219
Video Super-ResolutiononMSU Video Super Resolution Benchmark: Detail Restoration
PSNR· 2018-09-01
27.33
best: 31.669 (VRT)
ESRGAN: Enhanced Super-Resolution Generative Adversarial Networks arXiv:1809.00219
Video Super-ResolutiononMSU Video Super Resolution Benchmark: Detail Restoration
SSIM· 2018-09-01
0.808
best: 0.902 (VRT)
ESRGAN: Enhanced Super-Resolution Generative Adversarial Networks arXiv:1809.00219
3D Object Super-ResolutiononFFHQ 256 x 256 - 4x upscaling
FID· 2018-09-01
166.36
best: 5.36 (HiFaceGAN)
ESRGAN: Enhanced Super-Resolution Generative Adversarial Networks arXiv:1809.00219
3D Object Super-ResolutiononFFHQ 256 x 256 - 4x upscaling
MS-SSIM· 2018-09-01
0.747
best: 0.971 (HiFaceGAN)
ESRGAN: Enhanced Super-Resolution Generative Adversarial Networks arXiv:1809.00219
3D Object Super-ResolutiononFFHQ 256 x 256 - 4x upscaling
PSNR· 2018-09-01
15.43
best: 28.65 (HiFaceGAN)
ESRGAN: Enhanced Super-Resolution Generative Adversarial Networks arXiv:1809.00219
3D Object Super-ResolutiononFFHQ 256 x 256 - 4x upscaling
SSIM· 2018-09-01
0.267
best: 0.816 (HiFaceGAN)
ESRGAN: Enhanced Super-Resolution Generative Adversarial Networks arXiv:1809.00219
3D Object Super-ResolutiononFFHQ 512 x 512 - 4x upscaling
FED· 2018-09-01
0.1107
best: 0.1416 (Super-FAN)
ESRGAN: Enhanced Super-Resolution Generative Adversarial Networks arXiv:1809.00219
3D Object Super-ResolutiononFFHQ 512 x 512 - 4x upscaling
LLE· 2018-09-01
2.261
best: 2.702 (WaveletCNN)
ESRGAN: Enhanced Super-Resolution Generative Adversarial Networks arXiv:1809.00219
3D Object Super-ResolutiononFFHQ 512 x 512 - 4x upscaling
MS-SSIM· 2018-09-01
0.935
best: 0.976 (UNET++)
ESRGAN: Enhanced Super-Resolution Generative Adversarial Networks arXiv:1809.00219
3D Object Super-ResolutiononFFHQ 512 x 512 - 4x upscaling
NIQE· 2018-09-01
6.984
best: 13.901 (SRFBN)
ESRGAN: Enhanced Super-Resolution Generative Adversarial Networks arXiv:1809.00219
3D Object Super-ResolutiononFFHQ 512 x 512 - 4x upscaling
PSNR· 2018-09-01
27.134
best: 30.824 (HiFaceGAN)
ESRGAN: Enhanced Super-Resolution Generative Adversarial Networks arXiv:1809.00219
3D Object Super-ResolutiononFFHQ 512 x 512 - 4x upscaling
SSIM· 2018-09-01
0.741
best: 0.859 (UNET++)
ESRGAN: Enhanced Super-Resolution Generative Adversarial Networks arXiv:1809.00219
3D Object Super-ResolutiononFFHQ 1024 x 1024 - 4x upscaling
FID· 2018-09-01
72.73
best: 1.978 (HiFaceGAN)
ESRGAN: Enhanced Super-Resolution Generative Adversarial Networks arXiv:1809.00219
3D Object Super-ResolutiononFFHQ 1024 x 1024 - 4x upscaling
MS-SSIM· 2018-09-01
0.782
best: 0.975 (HiFaceGAN)
ESRGAN: Enhanced Super-Resolution Generative Adversarial Networks arXiv:1809.00219
3D Object Super-ResolutiononFFHQ 1024 x 1024 - 4x upscaling
PSNR· 2018-09-01
19.84
best: 34.1 (CAGFace)
ESRGAN: Enhanced Super-Resolution Generative Adversarial Networks arXiv:1809.00219
3D Object Super-ResolutiononFFHQ 1024 x 1024 - 4x upscaling
SSIM· 2018-09-01
0.353
best: 0.906 (CAGFace)
ESRGAN: Enhanced Super-Resolution Generative Adversarial Networks arXiv:1809.00219
3D Object Super-ResolutiononPIRM-test
NIQE· 2018-09-01
2.55
best: 2.71 (SRGAN)
ESRGAN: Enhanced Super-Resolution Generative Adversarial Networks arXiv:1809.00219
3D Object Super-ResolutiononMSU Video Upscalers: Quality Enhancement
PSNR· 2018-09-01
27.29
best: 31.28 (VEAI-AHQ-12)
ESRGAN: Enhanced Super-Resolution Generative Adversarial Networks arXiv:1809.00219
3D Object Super-ResolutiononMSU Video Upscalers: Quality Enhancement
SSIM· 2018-09-01
0.936
best: 0.939 (iSeeBetter)
ESRGAN: Enhanced Super-Resolution Generative Adversarial Networks arXiv:1809.00219
3D Object Super-ResolutiononMSU Video Upscalers: Quality Enhancement
VMAF· 2018-09-01
56.69
best: 61.2 (TecoGAN)
ESRGAN: Enhanced Super-Resolution Generative Adversarial Networks arXiv:1809.00219
3D Object Super-ResolutiononMSU Video Super Resolution Benchmark: Detail Restoration
1 - LPIPS· 2018-09-01
0.948
best: 0.623 (DFDnet)
ESRGAN: Enhanced Super-Resolution Generative Adversarial Networks arXiv:1809.00219
3D Object Super-ResolutiononMSU Video Super Resolution Benchmark: Detail Restoration
FPS· 2018-09-01
1.004
best: 5.882 (SRMD)
ESRGAN: Enhanced Super-Resolution Generative Adversarial Networks arXiv:1809.00219
3D Object Super-ResolutiononMSU Video Super Resolution Benchmark: Detail Restoration
PSNR· 2018-09-01
27.33
best: 31.669 (VRT)
ESRGAN: Enhanced Super-Resolution Generative Adversarial Networks arXiv:1809.00219
3D Object Super-ResolutiononMSU Video Super Resolution Benchmark: Detail Restoration
SSIM· 2018-09-01
0.808
best: 0.902 (VRT)
ESRGAN: Enhanced Super-Resolution Generative Adversarial Networks arXiv:1809.00219
3D Human Pose EstimationonMSU Video Super Resolution Benchmark: Detail Restoration
QRCRv1.0
0
best: 0.722 (VRT)
VideoonMSU Video Super Resolution Benchmark: Detail Restoration
QRCRv1.0
0
best: 0.722 (VRT)
Pose EstimationonMSU Video Super Resolution Benchmark: Detail Restoration
QRCRv1.0
0
best: 0.722 (VRT)
Video Super-ResolutiononMSU Video Super Resolution Benchmark: Detail Restoration
QRCRv1.0
0
best: 0.722 (VRT)
3D Object Super-ResolutiononMSU Video Super Resolution Benchmark: Detail Restoration
QRCRv1.0
0
best: 0.722 (VRT)

Methodology32 results

3DonMSU Video Super Resolution Benchmark: Detail Restoration
ERQAv1.0· 2018-09-01
0.735
best: 0.758 (VRT)
SOTA
ESRGAN: Enhanced Super-Resolution Generative Adversarial Networks arXiv:1809.00219
3DonMSU Video Super Resolution Benchmark: Detail Restoration
Subjective score· 2018-09-01
5.353
best: 7.628 (VRT)
SOTA
ESRGAN: Enhanced Super-Resolution Generative Adversarial Networks arXiv:1809.00219
3DonFFHQ 512 x 512 - 16x upscaling
FID· 2018-09-01
50.901
best: 11.389 (HiFaceGAN)
SOTA
ESRGAN: Enhanced Super-Resolution Generative Adversarial Networks arXiv:1809.00219
3DonFFHQ 512 x 512 - 16x upscaling
LPIPS· 2018-09-01
0.3928
best: 0.2449 (HiFaceGAN)
SOTA
ESRGAN: Enhanced Super-Resolution Generative Adversarial Networks arXiv:1809.00219
3DonFFHQ 512 x 512 - 16x upscaling
NIQE· 2018-09-01
15.383
SOTA
ESRGAN: Enhanced Super-Resolution Generative Adversarial Networks arXiv:1809.00219
16konFFHQ 512 x 512 - 4x upscaling
FID· 2018-09-01
3.503
best: 1.898 (HiFaceGAN)
SOTA
ESRGAN: Enhanced Super-Resolution Generative Adversarial Networks arXiv:1809.00219
16konFFHQ 512 x 512 - 4x upscaling
LPIPS· 2018-09-01
0.1221
best: 0.0723 (HiFaceGAN)
SOTA
ESRGAN: Enhanced Super-Resolution Generative Adversarial Networks arXiv:1809.00219
16konDIV2K val - 4x upscaling
PSNR· uses extra data· 2023-03-29
26.22
best: 29.25 (EDSR)
Implicit Diffusion Models for Continuous Super-Resolution arXiv:2303.16491
16konDIV2K val - 4x upscaling
SSIM· uses extra data· 2023-03-29
0.75
best: 0.9017 (EDSR)
Implicit Diffusion Models for Continuous Super-Resolution arXiv:2303.16491
3DonMSU Video Upscalers: Quality Enhancement
PSNR· 2018-09-01
27.29
best: 31.28 (VEAI-AHQ-12)
ESRGAN: Enhanced Super-Resolution Generative Adversarial Networks arXiv:1809.00219
3DonMSU Video Upscalers: Quality Enhancement
SSIM· 2018-09-01
0.936
best: 0.939 (iSeeBetter)
ESRGAN: Enhanced Super-Resolution Generative Adversarial Networks arXiv:1809.00219
3DonMSU Video Upscalers: Quality Enhancement
VMAF· 2018-09-01
56.69
best: 61.2 (TecoGAN)
ESRGAN: Enhanced Super-Resolution Generative Adversarial Networks arXiv:1809.00219
3DonMSU Video Super Resolution Benchmark: Detail Restoration
1 - LPIPS· 2018-09-01
0.948
best: 0.623 (DFDnet)
ESRGAN: Enhanced Super-Resolution Generative Adversarial Networks arXiv:1809.00219
3DonMSU Video Super Resolution Benchmark: Detail Restoration
FPS· 2018-09-01
1.004
best: 5.882 (SRMD)
ESRGAN: Enhanced Super-Resolution Generative Adversarial Networks arXiv:1809.00219
3DonMSU Video Super Resolution Benchmark: Detail Restoration
PSNR· 2018-09-01
27.33
best: 31.669 (VRT)
ESRGAN: Enhanced Super-Resolution Generative Adversarial Networks arXiv:1809.00219
3DonMSU Video Super Resolution Benchmark: Detail Restoration
SSIM· 2018-09-01
0.808
best: 0.902 (VRT)
ESRGAN: Enhanced Super-Resolution Generative Adversarial Networks arXiv:1809.00219
16konFFHQ 256 x 256 - 4x upscaling
FID· 2018-09-01
166.36
best: 5.36 (HiFaceGAN)
ESRGAN: Enhanced Super-Resolution Generative Adversarial Networks arXiv:1809.00219
16konFFHQ 256 x 256 - 4x upscaling
MS-SSIM· 2018-09-01
0.747
best: 0.971 (HiFaceGAN)
ESRGAN: Enhanced Super-Resolution Generative Adversarial Networks arXiv:1809.00219
16konFFHQ 256 x 256 - 4x upscaling
PSNR· 2018-09-01
15.43
best: 28.65 (HiFaceGAN)
ESRGAN: Enhanced Super-Resolution Generative Adversarial Networks arXiv:1809.00219
16konFFHQ 256 x 256 - 4x upscaling
SSIM· 2018-09-01
0.267
best: 0.816 (HiFaceGAN)
ESRGAN: Enhanced Super-Resolution Generative Adversarial Networks arXiv:1809.00219
16konFFHQ 512 x 512 - 4x upscaling
FED· 2018-09-01
0.1107
best: 0.1416 (Super-FAN)
ESRGAN: Enhanced Super-Resolution Generative Adversarial Networks arXiv:1809.00219
16konFFHQ 512 x 512 - 4x upscaling
LLE· 2018-09-01
2.261
best: 2.702 (WaveletCNN)
ESRGAN: Enhanced Super-Resolution Generative Adversarial Networks arXiv:1809.00219
16konFFHQ 512 x 512 - 4x upscaling
MS-SSIM· 2018-09-01
0.935
best: 0.976 (UNET++)
ESRGAN: Enhanced Super-Resolution Generative Adversarial Networks arXiv:1809.00219
16konFFHQ 512 x 512 - 4x upscaling
NIQE· 2018-09-01
6.984
best: 13.901 (SRFBN)
ESRGAN: Enhanced Super-Resolution Generative Adversarial Networks arXiv:1809.00219
16konFFHQ 512 x 512 - 4x upscaling
PSNR· 2018-09-01
27.134
best: 30.824 (HiFaceGAN)
ESRGAN: Enhanced Super-Resolution Generative Adversarial Networks arXiv:1809.00219
16konFFHQ 512 x 512 - 4x upscaling
SSIM· 2018-09-01
0.741
best: 0.859 (UNET++)
ESRGAN: Enhanced Super-Resolution Generative Adversarial Networks arXiv:1809.00219
16konFFHQ 1024 x 1024 - 4x upscaling
FID· 2018-09-01
72.73
best: 1.978 (HiFaceGAN)
ESRGAN: Enhanced Super-Resolution Generative Adversarial Networks arXiv:1809.00219
16konFFHQ 1024 x 1024 - 4x upscaling
MS-SSIM· 2018-09-01
0.782
best: 0.975 (HiFaceGAN)
ESRGAN: Enhanced Super-Resolution Generative Adversarial Networks arXiv:1809.00219
16konFFHQ 1024 x 1024 - 4x upscaling
PSNR· 2018-09-01
19.84
best: 34.1 (CAGFace)
ESRGAN: Enhanced Super-Resolution Generative Adversarial Networks arXiv:1809.00219
16konFFHQ 1024 x 1024 - 4x upscaling
SSIM· 2018-09-01
0.353
best: 0.906 (CAGFace)
ESRGAN: Enhanced Super-Resolution Generative Adversarial Networks arXiv:1809.00219
16konPIRM-test
NIQE· 2018-09-01
2.55
best: 2.71 (SRGAN)
ESRGAN: Enhanced Super-Resolution Generative Adversarial Networks arXiv:1809.00219
3DonMSU Video Super Resolution Benchmark: Detail Restoration
QRCRv1.0
0
best: 0.722 (VRT)

Graphs29 results

Super-ResolutiononFFHQ 512 x 512 - 4x upscaling
FID· 2018-09-01
3.503
best: 1.898 (HiFaceGAN)
SOTA
ESRGAN: Enhanced Super-Resolution Generative Adversarial Networks arXiv:1809.00219
Super-ResolutiononFFHQ 512 x 512 - 4x upscaling
LPIPS· 2018-09-01
0.1221
best: 0.0723 (HiFaceGAN)
SOTA
ESRGAN: Enhanced Super-Resolution Generative Adversarial Networks arXiv:1809.00219
Super-ResolutiononMSU Video Super Resolution Benchmark: Detail Restoration
ERQAv1.0· 2018-09-01
0.735
best: 0.758 (VRT)
SOTA
ESRGAN: Enhanced Super-Resolution Generative Adversarial Networks arXiv:1809.00219
Super-ResolutiononMSU Video Super Resolution Benchmark: Detail Restoration
Subjective score· 2018-09-01
5.353
best: 7.628 (VRT)
SOTA
ESRGAN: Enhanced Super-Resolution Generative Adversarial Networks arXiv:1809.00219
Super-ResolutiononDIV2K val - 4x upscaling
PSNR· uses extra data· 2023-03-29
26.22
best: 29.25 (EDSR)
Implicit Diffusion Models for Continuous Super-Resolution arXiv:2303.16491
Super-ResolutiononDIV2K val - 4x upscaling
SSIM· uses extra data· 2023-03-29
0.75
best: 0.9017 (EDSR)
Implicit Diffusion Models for Continuous Super-Resolution arXiv:2303.16491
Super-ResolutiononFFHQ 256 x 256 - 4x upscaling
FID· 2018-09-01
166.36
best: 5.36 (HiFaceGAN)
ESRGAN: Enhanced Super-Resolution Generative Adversarial Networks arXiv:1809.00219
Super-ResolutiononFFHQ 256 x 256 - 4x upscaling
MS-SSIM· 2018-09-01
0.747
best: 0.971 (HiFaceGAN)
ESRGAN: Enhanced Super-Resolution Generative Adversarial Networks arXiv:1809.00219
Super-ResolutiononFFHQ 256 x 256 - 4x upscaling
PSNR· 2018-09-01
15.43
best: 28.65 (HiFaceGAN)
ESRGAN: Enhanced Super-Resolution Generative Adversarial Networks arXiv:1809.00219
Super-ResolutiononFFHQ 256 x 256 - 4x upscaling
SSIM· 2018-09-01
0.267
best: 0.816 (HiFaceGAN)
ESRGAN: Enhanced Super-Resolution Generative Adversarial Networks arXiv:1809.00219
Super-ResolutiononFFHQ 512 x 512 - 4x upscaling
FED· 2018-09-01
0.1107
best: 0.1416 (Super-FAN)
ESRGAN: Enhanced Super-Resolution Generative Adversarial Networks arXiv:1809.00219
Super-ResolutiononFFHQ 512 x 512 - 4x upscaling
LLE· 2018-09-01
2.261
best: 2.702 (WaveletCNN)
ESRGAN: Enhanced Super-Resolution Generative Adversarial Networks arXiv:1809.00219
Super-ResolutiononFFHQ 512 x 512 - 4x upscaling
MS-SSIM· 2018-09-01
0.935
best: 0.976 (UNET++)
ESRGAN: Enhanced Super-Resolution Generative Adversarial Networks arXiv:1809.00219
Super-ResolutiononFFHQ 512 x 512 - 4x upscaling
NIQE· 2018-09-01
6.984
best: 13.901 (SRFBN)
ESRGAN: Enhanced Super-Resolution Generative Adversarial Networks arXiv:1809.00219
Super-ResolutiononFFHQ 512 x 512 - 4x upscaling
PSNR· 2018-09-01
27.134
best: 30.824 (HiFaceGAN)
ESRGAN: Enhanced Super-Resolution Generative Adversarial Networks arXiv:1809.00219
Super-ResolutiononFFHQ 512 x 512 - 4x upscaling
SSIM· 2018-09-01
0.741
best: 0.859 (UNET++)
ESRGAN: Enhanced Super-Resolution Generative Adversarial Networks arXiv:1809.00219
Super-ResolutiononFFHQ 1024 x 1024 - 4x upscaling
FID· 2018-09-01
72.73
best: 1.978 (HiFaceGAN)
ESRGAN: Enhanced Super-Resolution Generative Adversarial Networks arXiv:1809.00219
Super-ResolutiononFFHQ 1024 x 1024 - 4x upscaling
MS-SSIM· 2018-09-01
0.782
best: 0.975 (HiFaceGAN)
ESRGAN: Enhanced Super-Resolution Generative Adversarial Networks arXiv:1809.00219
Super-ResolutiononFFHQ 1024 x 1024 - 4x upscaling
PSNR· 2018-09-01
19.84
best: 34.1 (CAGFace)
ESRGAN: Enhanced Super-Resolution Generative Adversarial Networks arXiv:1809.00219
Super-ResolutiononFFHQ 1024 x 1024 - 4x upscaling
SSIM· 2018-09-01
0.353
best: 0.906 (CAGFace)
ESRGAN: Enhanced Super-Resolution Generative Adversarial Networks arXiv:1809.00219
Super-ResolutiononPIRM-test
NIQE· 2018-09-01
2.55
best: 2.71 (SRGAN)
ESRGAN: Enhanced Super-Resolution Generative Adversarial Networks arXiv:1809.00219
Super-ResolutiononMSU Video Upscalers: Quality Enhancement
PSNR· 2018-09-01
27.29
best: 31.28 (VEAI-AHQ-12)
ESRGAN: Enhanced Super-Resolution Generative Adversarial Networks arXiv:1809.00219
Super-ResolutiononMSU Video Upscalers: Quality Enhancement
SSIM· 2018-09-01
0.936
best: 0.939 (iSeeBetter)
ESRGAN: Enhanced Super-Resolution Generative Adversarial Networks arXiv:1809.00219
Super-ResolutiononMSU Video Upscalers: Quality Enhancement
VMAF· 2018-09-01
56.69
best: 61.2 (TecoGAN)
ESRGAN: Enhanced Super-Resolution Generative Adversarial Networks arXiv:1809.00219
Super-ResolutiononMSU Video Super Resolution Benchmark: Detail Restoration
1 - LPIPS· 2018-09-01
0.948
best: 0.623 (DFDnet)
ESRGAN: Enhanced Super-Resolution Generative Adversarial Networks arXiv:1809.00219
Super-ResolutiononMSU Video Super Resolution Benchmark: Detail Restoration
FPS· 2018-09-01
1.004
best: 5.882 (SRMD)
ESRGAN: Enhanced Super-Resolution Generative Adversarial Networks arXiv:1809.00219
Super-ResolutiononMSU Video Super Resolution Benchmark: Detail Restoration
PSNR· 2018-09-01
27.33
best: 31.669 (VRT)
ESRGAN: Enhanced Super-Resolution Generative Adversarial Networks arXiv:1809.00219
Super-ResolutiononMSU Video Super Resolution Benchmark: Detail Restoration
SSIM· 2018-09-01
0.808
best: 0.902 (VRT)
ESRGAN: Enhanced Super-Resolution Generative Adversarial Networks arXiv:1809.00219
Super-ResolutiononMSU Video Super Resolution Benchmark: Detail Restoration
QRCRv1.0
0
best: 0.722 (VRT)

Knowledge Base20 results

2D Human Pose EstimationonMSU Video Super Resolution Benchmark: Detail Restoration
ERQAv1.0· 2018-09-01
0.735
best: 0.758 (VRT)
SOTA
ESRGAN: Enhanced Super-Resolution Generative Adversarial Networks arXiv:1809.00219
2D Human Pose EstimationonMSU Video Super Resolution Benchmark: Detail Restoration
Subjective score· 2018-09-01
5.353
best: 7.628 (VRT)
SOTA
ESRGAN: Enhanced Super-Resolution Generative Adversarial Networks arXiv:1809.00219
3D Absolute Human Pose EstimationonMSU Video Super Resolution Benchmark: Detail Restoration
ERQAv1.0· 2018-09-01
0.735
best: 0.758 (VRT)
SOTA
ESRGAN: Enhanced Super-Resolution Generative Adversarial Networks arXiv:1809.00219
3D Absolute Human Pose EstimationonMSU Video Super Resolution Benchmark: Detail Restoration
Subjective score· 2018-09-01
5.353
best: 7.628 (VRT)
SOTA
ESRGAN: Enhanced Super-Resolution Generative Adversarial Networks arXiv:1809.00219
2D Human Pose EstimationonMSU Video Upscalers: Quality Enhancement
PSNR· 2018-09-01
27.29
best: 31.28 (VEAI-AHQ-12)
ESRGAN: Enhanced Super-Resolution Generative Adversarial Networks arXiv:1809.00219
2D Human Pose EstimationonMSU Video Upscalers: Quality Enhancement
SSIM· 2018-09-01
0.936
best: 0.939 (iSeeBetter)
ESRGAN: Enhanced Super-Resolution Generative Adversarial Networks arXiv:1809.00219
2D Human Pose EstimationonMSU Video Upscalers: Quality Enhancement
VMAF· 2018-09-01
56.69
best: 61.2 (TecoGAN)
ESRGAN: Enhanced Super-Resolution Generative Adversarial Networks arXiv:1809.00219
2D Human Pose EstimationonMSU Video Super Resolution Benchmark: Detail Restoration
1 - LPIPS· 2018-09-01
0.948
best: 0.623 (DFDnet)
ESRGAN: Enhanced Super-Resolution Generative Adversarial Networks arXiv:1809.00219
2D Human Pose EstimationonMSU Video Super Resolution Benchmark: Detail Restoration
FPS· 2018-09-01
1.004
best: 5.882 (SRMD)
ESRGAN: Enhanced Super-Resolution Generative Adversarial Networks arXiv:1809.00219
2D Human Pose EstimationonMSU Video Super Resolution Benchmark: Detail Restoration
PSNR· 2018-09-01
27.33
best: 31.669 (VRT)
ESRGAN: Enhanced Super-Resolution Generative Adversarial Networks arXiv:1809.00219
2D Human Pose EstimationonMSU Video Super Resolution Benchmark: Detail Restoration
SSIM· 2018-09-01
0.808
best: 0.902 (VRT)
ESRGAN: Enhanced Super-Resolution Generative Adversarial Networks arXiv:1809.00219
3D Absolute Human Pose EstimationonMSU Video Upscalers: Quality Enhancement
PSNR· 2018-09-01
27.29
best: 31.28 (VEAI-AHQ-12)
ESRGAN: Enhanced Super-Resolution Generative Adversarial Networks arXiv:1809.00219
3D Absolute Human Pose EstimationonMSU Video Upscalers: Quality Enhancement
SSIM· 2018-09-01
0.936
best: 0.939 (iSeeBetter)
ESRGAN: Enhanced Super-Resolution Generative Adversarial Networks arXiv:1809.00219
3D Absolute Human Pose EstimationonMSU Video Upscalers: Quality Enhancement
VMAF· 2018-09-01
56.69
best: 61.2 (TecoGAN)
ESRGAN: Enhanced Super-Resolution Generative Adversarial Networks arXiv:1809.00219
3D Absolute Human Pose EstimationonMSU Video Super Resolution Benchmark: Detail Restoration
1 - LPIPS· 2018-09-01
0.948
best: 0.623 (DFDnet)
ESRGAN: Enhanced Super-Resolution Generative Adversarial Networks arXiv:1809.00219
3D Absolute Human Pose EstimationonMSU Video Super Resolution Benchmark: Detail Restoration
FPS· 2018-09-01
1.004
best: 5.882 (SRMD)
ESRGAN: Enhanced Super-Resolution Generative Adversarial Networks arXiv:1809.00219
3D Absolute Human Pose EstimationonMSU Video Super Resolution Benchmark: Detail Restoration
PSNR· 2018-09-01
27.33
best: 31.669 (VRT)
ESRGAN: Enhanced Super-Resolution Generative Adversarial Networks arXiv:1809.00219
3D Absolute Human Pose EstimationonMSU Video Super Resolution Benchmark: Detail Restoration
SSIM· 2018-09-01
0.808
best: 0.902 (VRT)
ESRGAN: Enhanced Super-Resolution Generative Adversarial Networks arXiv:1809.00219
2D Human Pose EstimationonMSU Video Super Resolution Benchmark: Detail Restoration
QRCRv1.0
0
best: 0.722 (VRT)
3D Absolute Human Pose EstimationonMSU Video Super Resolution Benchmark: Detail Restoration
QRCRv1.0
0
best: 0.722 (VRT)

Playing Games10 results

3D Face AnimationonMSU Video Super Resolution Benchmark: Detail Restoration
ERQAv1.0· 2018-09-01
0.735
best: 0.758 (VRT)
SOTA
ESRGAN: Enhanced Super-Resolution Generative Adversarial Networks arXiv:1809.00219
3D Face AnimationonMSU Video Super Resolution Benchmark: Detail Restoration
Subjective score· 2018-09-01
5.353
best: 7.628 (VRT)
SOTA
ESRGAN: Enhanced Super-Resolution Generative Adversarial Networks arXiv:1809.00219
3D Face AnimationonMSU Video Upscalers: Quality Enhancement
PSNR· 2018-09-01
27.29
best: 31.28 (VEAI-AHQ-12)
ESRGAN: Enhanced Super-Resolution Generative Adversarial Networks arXiv:1809.00219
3D Face AnimationonMSU Video Upscalers: Quality Enhancement
SSIM· 2018-09-01
0.936
best: 0.939 (iSeeBetter)
ESRGAN: Enhanced Super-Resolution Generative Adversarial Networks arXiv:1809.00219
3D Face AnimationonMSU Video Upscalers: Quality Enhancement
VMAF· 2018-09-01
56.69
best: 61.2 (TecoGAN)
ESRGAN: Enhanced Super-Resolution Generative Adversarial Networks arXiv:1809.00219
3D Face AnimationonMSU Video Super Resolution Benchmark: Detail Restoration
1 - LPIPS· 2018-09-01
0.948
best: 0.623 (DFDnet)
ESRGAN: Enhanced Super-Resolution Generative Adversarial Networks arXiv:1809.00219
3D Face AnimationonMSU Video Super Resolution Benchmark: Detail Restoration
FPS· 2018-09-01
1.004
best: 5.882 (SRMD)
ESRGAN: Enhanced Super-Resolution Generative Adversarial Networks arXiv:1809.00219
3D Face AnimationonMSU Video Super Resolution Benchmark: Detail Restoration
PSNR· 2018-09-01
27.33
best: 31.669 (VRT)
ESRGAN: Enhanced Super-Resolution Generative Adversarial Networks arXiv:1809.00219
3D Face AnimationonMSU Video Super Resolution Benchmark: Detail Restoration
SSIM· 2018-09-01
0.808
best: 0.902 (VRT)
ESRGAN: Enhanced Super-Resolution Generative Adversarial Networks arXiv:1809.00219
3D Face AnimationonMSU Video Super Resolution Benchmark: Detail Restoration
QRCRv1.0
0
best: 0.722 (VRT)

Audio10 results

1 Image, 2*2 StitchionMSU Video Super Resolution Benchmark: Detail Restoration
ERQAv1.0· 2018-09-01
0.735
best: 0.758 (VRT)
SOTA
ESRGAN: Enhanced Super-Resolution Generative Adversarial Networks arXiv:1809.00219
1 Image, 2*2 StitchionMSU Video Super Resolution Benchmark: Detail Restoration
Subjective score· 2018-09-01
5.353
best: 7.628 (VRT)
SOTA
ESRGAN: Enhanced Super-Resolution Generative Adversarial Networks arXiv:1809.00219
1 Image, 2*2 StitchionMSU Video Upscalers: Quality Enhancement
PSNR· 2018-09-01
27.29
best: 31.28 (VEAI-AHQ-12)
ESRGAN: Enhanced Super-Resolution Generative Adversarial Networks arXiv:1809.00219
1 Image, 2*2 StitchionMSU Video Upscalers: Quality Enhancement
SSIM· 2018-09-01
0.936
best: 0.939 (iSeeBetter)
ESRGAN: Enhanced Super-Resolution Generative Adversarial Networks arXiv:1809.00219
1 Image, 2*2 StitchionMSU Video Upscalers: Quality Enhancement
VMAF· 2018-09-01
56.69
best: 61.2 (TecoGAN)
ESRGAN: Enhanced Super-Resolution Generative Adversarial Networks arXiv:1809.00219
1 Image, 2*2 StitchionMSU Video Super Resolution Benchmark: Detail Restoration
1 - LPIPS· 2018-09-01
0.948
best: 0.623 (DFDnet)
ESRGAN: Enhanced Super-Resolution Generative Adversarial Networks arXiv:1809.00219
1 Image, 2*2 StitchionMSU Video Super Resolution Benchmark: Detail Restoration
FPS· 2018-09-01
1.004
best: 5.882 (SRMD)
ESRGAN: Enhanced Super-Resolution Generative Adversarial Networks arXiv:1809.00219
1 Image, 2*2 StitchionMSU Video Super Resolution Benchmark: Detail Restoration
PSNR· 2018-09-01
27.33
best: 31.669 (VRT)
ESRGAN: Enhanced Super-Resolution Generative Adversarial Networks arXiv:1809.00219
1 Image, 2*2 StitchionMSU Video Super Resolution Benchmark: Detail Restoration
SSIM· 2018-09-01
0.808
best: 0.902 (VRT)
ESRGAN: Enhanced Super-Resolution Generative Adversarial Networks arXiv:1809.00219
1 Image, 2*2 StitchionMSU Video Super Resolution Benchmark: Detail Restoration
QRCRv1.0
0
best: 0.722 (VRT)

Music3 results

Facial Recognition and ModellingonFFHQ 512 x 512 - 16x upscaling
FID· 2018-09-01
50.901
best: 11.389 (HiFaceGAN)
SOTA
ESRGAN: Enhanced Super-Resolution Generative Adversarial Networks arXiv:1809.00219
Facial Recognition and ModellingonFFHQ 512 x 512 - 16x upscaling
LPIPS· 2018-09-01
0.3928
best: 0.2449 (HiFaceGAN)
SOTA
ESRGAN: Enhanced Super-Resolution Generative Adversarial Networks arXiv:1809.00219
Facial Recognition and ModellingonFFHQ 512 x 512 - 16x upscaling
NIQE· 2018-09-01
15.383
SOTA
ESRGAN: Enhanced Super-Resolution Generative Adversarial Networks arXiv:1809.00219

Medical3 results

3D Face ModellingonFFHQ 512 x 512 - 16x upscaling
FID· 2018-09-01
50.901
best: 11.389 (HiFaceGAN)
SOTA
ESRGAN: Enhanced Super-Resolution Generative Adversarial Networks arXiv:1809.00219
3D Face ModellingonFFHQ 512 x 512 - 16x upscaling
LPIPS· 2018-09-01
0.3928
best: 0.2449 (HiFaceGAN)
SOTA
ESRGAN: Enhanced Super-Resolution Generative Adversarial Networks arXiv:1809.00219
3D Face ModellingonFFHQ 512 x 512 - 16x upscaling
NIQE· 2018-09-01
15.383
SOTA
ESRGAN: Enhanced Super-Resolution Generative Adversarial Networks arXiv:1809.00219