Papers › ESRGAN: Enhanced Super-Resolution Generative Adversarial Networks

ESRGAN: Enhanced Super-Resolution Generative Adversarial Networks

1 Sep 2018arXiv:1809.00219archive 2025-07-28

Xintao Wang, Ke Yu, Shixiang Wu, Jinjin Gu, Yihao Liu, Chao Dong, Chen Change Loy, Yu Qiao, Xiaoou Tang

The Super-Resolution Generative Adversarial Network (SRGAN) is a seminal work that is capable of generating realistic textures during single image super-resolution. However, the hallucinated details are often accompanied with unpleasant artifacts. To further enhance the visual quality, we thoroughly study three key components of SRGAN - network architecture, adversarial loss and perceptual loss, and improve each of them to derive an Enhanced SRGAN (ESRGAN). In particular, we introduce the Residual-in-Residual Dense Block (RRDB) without batch normalization as the basic network building unit. Moreover, we borrow the idea from relativistic GAN to let the discriminator predict relative realness instead of the absolute value. Finally, we improve the perceptual loss by using the features before activation, which could provide stronger supervision for brightness consistency and texture recovery. Benefiting from these improvements, the proposed ESRGAN achieves consistently better visual quality with more realistic and natural textures than SRGAN and won the first place in the PIRM2018-SR Challenge. The code is available at https://github.com/xinntao/ESRGAN .

PaperPDFCodeCode Syntology ran

In Syntology Open this paper in Syntology's Atlas, the map of the papers in Syntology's graph and their citations.

For agents, Syntology's MCP tool lists every function and class Syntology harvested from this paper and whether it ran (how to connect): get_harvested_code_for_paper(arxiv_id="1809.00219")

Code

Syntology Ran 8 of 44 code samples harvested from 7 repositories linked to this paper; 36 have no recorded run. Of those that ran: 4 ran · our draft was wrong; 4 ran with no contract checked.

By repository: official repository: 1 sample from 1 repository, 1 ran; community (archive-listed): 43 samples from 6 repositories, 7 ran. The run record, sample by sample. “Ran” means executed on a synthesized input, not that the code is correct or reproduces the paper.

46 repositories listed; official and paper-mentioned ones first.

xinntao/ESRGAN officialmentioned in papermentioned on GitHubpytorchApache-2.0 report
AnjanaGJoseph/Super-Resolution-GAN mentioned on GitHubGPL-3.0 report
CorentinMAG/SRGAN mentioned on GitHub report
Dictator890/ESRGAN-Android-Demo mentioned on GitHubApache-2.0 report
FirstPotatoMan/TensorflowJS_ESRGAN mentioned on GitHubtfMIT report
HyeongJu916/Boaz-SR-ESRGAN-PyTorch mentioned on GitHubpytorchApache-2.0 report
Lornatang/ESRGAN-PyTorch mentioned on GitHubpytorchApache-2.0 report
RezisEwig/esrgan mentioned on GitHubtf report
YanyingZH/srgan-esrgan-pytorch mentioned on GitHubpytorch report
YvanG/VQGAN-CLIP mentioned on GitHubpytorch report
aba450/Super-Resolution mentioned on GitHubpytorch report
alililia/ascend_ESRGAN mentioned on GitHubmindsporeApache-2.0 report
allenai/satlas-super-resolution mentioned on GitHubpytorchApache-2.0 report
amogh7joshi/media-vision mentioned on GitHubpytorch report
cviaai/PSF-GAN-ESTIMATION mentioned on GitHubpytorchMIT report
doantienthongbku/ESRGAN mentioned on GitHubpytorch report
eriklindernoren/PyTorch-GAN mentioned on GitHubpytorch report
fenghansen/ESRGAN-Keras mentioned on GitHubtf report
idealo/image-super-resolution mentioned on GitHubtf report
itsuki8914/ESRGAN-TensorFlow mentioned on GitHubtf report
kingcong/gpu_ESRGAN mentioned on GitHubmindsporenot reachable when probed 2026-09-18 — repositories for recent papers often appear after camera-ready report
luciennnnnnn/dualformer mentioned on GitHubpytorch report
nabarunbaruaAIML/ATCC_Yolov5 mentioned on GitHubpytorch report
nannau/DoWnGAN mentioned on GitHubpytorchGPL-3.0 report
peteryuX/esrgan-tf2 mentioned on GitHubtfMIT report
sberbank-ai/DigiTales mentioned on GitHubtf report
sberbank-ai/DigiTeller mentioned on GitHubtf report
sberbank-ai/real-esrgan mentioned on GitHubpytorchBSD-3-Clause report
sdauzcm/sr-basicsr mentioned on GitHubpytorchApache-2.0 report
u7javed/Image-Enhancer-via-ESRGAN mentioned on GitHubpytorchMIT report
yangyucheng000/ESRGAN mentioned on GitHubmindspore report

Repository list and official/mentioned flags are the archive's, frozen 2025-07-28. Reachability, where shown, is from one Syntology probe window (2026-09-16 to 2026-09-18); repositories not probed show nothing. GitHub stars are not tracked.

Code Syntology ran Syntology

44 samples harvested; 8 ran; 0 honoured the contract we drafted; 36 have no recorded run. Read from Syntology's graph 2026-09-24; that is when this build read the record, not when the samples ran.

4ran · our draft was wrong
4ran
36unverified

Licence: 0 of the 44 samples are pointer only, meaning Syntology does not serve that copy's text. This page shows no code text for any sample; each one links to its file in the repository.

Harvested from 7 repositories linked to this paper, official or community; each sample names its own and says which. “Ran” means the sample executed on a synthesized input. It does not mean the output is correct, and nothing here reproduces the paper's results. “Honoured” and “violated” refer to a contract Syntology drafted from the code itself; “our draft was wrong” and “fixture could not drive it” are failures of Syntology's instrument, not of the code.

Each sample ends with its code_sha256, Syntology's identity for that exact code. An agent fetches the stored sample with Syntology's MCP tool get_code(code_sha256="…") (how to connect); click an identity to copy that call.

Repository labels, per sample. official repository: The archive marks this repository official for the paper. named in the paper: The archive records that the paper mentions this repository; it is not marked official. community (archive-listed): In the archive's code links for this paper, not marked official and not recorded as mentioned in the paper. found in paper text by Syntology: Syntology found this repository in the paper's own text; whether it is the authors' implementation is not asserted. community: Not in the archive's code links for this paper; a community repository Syntology harvested. Samples from a repository marked official are listed first. Licence labels name the repository's licence as recorded at harvest. “Pointer only” means Syntology does not serve that copy's text, for one of four reasons: no licence file was found; the licence was not identified; the licence is recorded as permissive but that copy's record is not marked cleared; or the licence is outside the permissive list Syntology serves text under (MIT, Apache-2.0, BSD and similar). Some licences outside that list permit redistribution, such as WTFPL, and GPL-3.0 under its conditions; they are simply not on the list. Hover a licence label for the reason. File links open the file on GitHub at the default branch, which may have changed since the harvest.

make_layer xinntao/ESRGAN/RRDBNet_arch.py official repository ran · our draft was wrong Apache-2.0 (permissive) · c04a23de0531da4a · report
act sdauzcm/sr-basicsr/codes/models/modules/block.py community (archive-listed) ran · our draft was wrong Apache-2.0 (permissive) · 15d829a065897df0 · report
calculate_psnr sdauzcm/sr-basicsr/metrics/calculate_PSNR_SSIM.py community (archive-listed) ran fingerprinted Apache-2.0 (permissive) · 9394ff2c25b8a988 · report
cubic HyeongJu916/Boaz-SR-ESRGAN-PyTorch/esrgan_pytorch/utils.py community (archive-listed) ran fingerprinted Apache-2.0 (permissive) · cf8b0785f1ca3da5 · report
flow_warp cviaai/PSF-GAN-ESTIMATION/mmsr_code/arch_util.py community (archive-listed) ran MIT (permissive) · 7018e324f2150c43 · report
norm sdauzcm/sr-basicsr/codes/models/modules/block.py community (archive-listed) ran · our draft was wrong Apache-2.0 (permissive) · b6a238bb90c0069b · report
pad sdauzcm/sr-basicsr/codes/models/modules/block.py community (archive-listed) ran · our draft was wrong Apache-2.0 (permissive) · 4a56ce1917d87f9b · report
ssim sdauzcm/sr-basicsr/metrics/calculate_PSNR_SSIM.py community (archive-listed) ran fingerprinted Apache-2.0 (permissive) · ba721aeacc165234 · report
ContentLoss peteryuX/esrgan-tf2/modules/losses.py community (archive-listed) unverified MIT (permissive) · a1eae53995164a47 · report
CosineAnnealingLR_Restart peteryuX/esrgan-tf2/modules/lr_scheduler.py community (archive-listed) unverified MIT (permissive) · 28087fef70cd7834 · report
DiscriminatorLoss peteryuX/esrgan-tf2/modules/losses.py community (archive-listed) unverified MIT (permissive) · 33f7fba25367dd03 · report
DiscriminatorVGG128 peteryuX/esrgan-tf2/modules/models.py community (archive-listed) unverified MIT (permissive) · 3c7dc4c98687dd3e · report
MultiStepLR peteryuX/esrgan-tf2/modules/lr_scheduler.py community (archive-listed) unverified MIT (permissive) · 9e0df5e5f87592dc · report
PixelLoss peteryuX/esrgan-tf2/modules/losses.py community (archive-listed) unverified MIT (permissive) · 3a8e26a89fb2bbca · report
RRDB_Model peteryuX/esrgan-tf2/modules/models.py community (archive-listed) unverified MIT (permissive) · af49e058cf9418d0 · report
aggd_features HyeongJu916/Boaz-SR-ESRGAN-PyTorch/esrgan_pytorch/calculate_niqe.py community (archive-listed) unverified Apache-2.0 (permissive) · 651043e835d45659 · report
cal_niqe HyeongJu916/Boaz-SR-ESRGAN-PyTorch/esrgan_pytorch/calculate_niqe.py community (archive-listed) unverified Apache-2.0 (permissive) · 6978b4d6b12aaa7f · report
calculate_ssim sdauzcm/sr-basicsr/metrics/calculate_PSNR_SSIM.py community (archive-listed) unverified Apache-2.0 (permissive) · 5aef99e58703c44a · report
calculate_weights_indices HyeongJu916/Boaz-SR-ESRGAN-PyTorch/esrgan_pytorch/utils.py community (archive-listed) unverified Apache-2.0 (permissive) · 4321ce86e7b65c96 · report
check_image_file HyeongJu916/Boaz-SR-ESRGAN-PyTorch/esrgan_pytorch/dataset.py community (archive-listed) unverified Apache-2.0 (permissive) · b97e4bafa6d5a536 · report
compute_image_mscn_transform HyeongJu916/Boaz-SR-ESRGAN-PyTorch/esrgan_pytorch/calculate_niqe.py community (archive-listed) unverified Apache-2.0 (permissive) · 668d7b024ce7ffae · report
create_lr_hr_pair peteryuX/esrgan-tf2/modules/utils.py community (archive-listed) unverified MIT (permissive) · 529169af10664b73 · report
define_loss Lornatang/ESRGAN-PyTorch/train_net.py community (archive-listed) unverified Apache-2.0 (permissive) · cc359b637efed011 · report
esrgan_4x4_16 HyeongJu916/Boaz-SR-ESRGAN-PyTorch/esrgan_pytorch/models/generator.py community (archive-listed) unverified Apache-2.0 (permissive) · 57b05189bf70b18d · report
esrgan_4x4_23 HyeongJu916/Boaz-SR-ESRGAN-PyTorch/esrgan_pytorch/models/generator.py community (archive-listed) unverified Apache-2.0 (permissive) · 204745c935fe8615 · report
image_to_tensor Lornatang/ESRGAN-PyTorch/imgproc.py community (archive-listed) unverified Apache-2.0 (permissive) · d14280e408db8305 · report
img_convert alililia/ascend_ESRGAN/postprocess.py community (archive-listed) unverified Apache-2.0 (permissive) · fcd6dda250c5738b · report
imresize HyeongJu916/Boaz-SR-ESRGAN-PyTorch/esrgan_pytorch/utils.py community (archive-listed) unverified Apache-2.0 (permissive) · 5dcd30c949329aa2 · report
inference HyeongJu916/Boaz-SR-ESRGAN-PyTorch/esrgan_pytorch/utils/common.py community (archive-listed) unverified Apache-2.0 (permissive) · bf876a2e4c5b05d9 · report
load_pretrained_state_dict Lornatang/ESRGAN-PyTorch/utils.py community (archive-listed) unverified Apache-2.0 (permissive) · e0364deba9b8c665 · report
load_state_dict Lornatang/ESRGAN-PyTorch/utils.py community (archive-listed) unverified Apache-2.0 (permissive) · 08e3b22c15672b97 · report
load_tfrecord_dataset peteryuX/esrgan-tf2/modules/dataset.py community (archive-listed) unverified MIT (permissive) · 27670b27fcb7619d · report
load_yaml peteryuX/esrgan-tf2/modules/utils.py community (archive-listed) unverified MIT (permissive) · c6d8e748cf074438 · report
niqe HyeongJu916/Boaz-SR-ESRGAN-PyTorch/esrgan_pytorch/utils/calculate_niqe.py community (archive-listed) unverified Apache-2.0 (permissive) · ef9c26bb9c8ac3a9 · report
padding alililia/ascend_ESRGAN/preprocess.py community (archive-listed) unverified Apache-2.0 (permissive) · e4f005d07cebcb69 · report
preprocess_one_image Lornatang/ESRGAN-PyTorch/imgproc.py community (archive-listed) unverified Apache-2.0 (permissive) · 096d345ecfa304ae · report
scandir alililia/ascend_ESRGAN/src/util/extract_subimages.py community (archive-listed) unverified Apache-2.0 (permissive) · 914623baf54d27d0 · report
select_device HyeongJu916/Boaz-SR-ESRGAN-PyTorch/esrgan_pytorch/utils/device.py community (archive-listed) unverified Apache-2.0 (permissive) · 247264d6a145b502 · report
tensor_to_image Lornatang/ESRGAN-PyTorch/imgproc.py community (archive-listed) unverified Apache-2.0 (permissive) · 00afe15c4a0c8764 · report
test_lpips HyeongJu916/Boaz-SR-ESRGAN-PyTorch/esrgan_pytorch/utils/estimate.py community (archive-listed) unverified Apache-2.0 (permissive) · 0369eca684c8dd34 · report
train_hr_transform HyeongJu916/Boaz-SR-ESRGAN-PyTorch/esrgan_pytorch/dataset.py community (archive-listed) unverified Apache-2.0 (permissive) · 5a0886f0454bfb05 · report
train_lr_transform HyeongJu916/Boaz-SR-ESRGAN-PyTorch/esrgan_pytorch/dataset.py community (archive-listed) unverified Apache-2.0 (permissive) · a73753ac253bbdd6 · report
unpadding alililia/ascend_ESRGAN/postprocess.py community (archive-listed) unverified Apache-2.0 (permissive) · a8e0c9457b927abb · report
worker alililia/ascend_ESRGAN/src/util/extract_subimages.py community (archive-listed) unverified Apache-2.0 (permissive) · 2304a4ee7dcb90c0 · report

Tasks

Face HallucinationImage Super-ResolutionSuper-ResolutionVideo Super-Resolution

1 archive task tag without a task page not shown.

Results from the paper archive 2025-07-28

TaskDatasetModelMetricValueRank at snapshotLeaderboardReport
Face Hallucination FFHQ 512 x 512 - 16x upscaling ESRGAN FID 50.901 #2 of 4 Archive leaderboard report
Face Hallucination FFHQ 512 x 512 - 16x upscaling ESRGAN LPIPS 0.3928 #2 of 4 Archive leaderboard report
Face Hallucination FFHQ 512 x 512 - 16x upscaling ESRGAN NIQE 15.383 #2 of 4 Archive leaderboard report
Image Super-Resolution BSD100 - 4x upscaling SRGAN + Residual-in-Residual Dense Block PSNR 27.85 #15 of 71 Archive leaderboard report
Image Super-Resolution BSD100 - 4x upscaling SRGAN + Residual-in-Residual Dense Block SSIM 0.7455 #15 of 71 Archive leaderboard report
Image Super-Resolution FFHQ 1024 x 1024 - 4x upscaling ESRGAN FID 72.73 #9 of 9 Archive leaderboard report
Image Super-Resolution FFHQ 1024 x 1024 - 4x upscaling ESRGAN MS-SSIM 0.782 #9 of 9 Archive leaderboard report
Image Super-Resolution FFHQ 1024 x 1024 - 4x upscaling ESRGAN PSNR 19.84 #9 of 9 Archive leaderboard report
Image Super-Resolution FFHQ 1024 x 1024 - 4x upscaling ESRGAN SSIM 0.353 #9 of 9 Archive leaderboard report
Image Super-Resolution FFHQ 256 x 256 - 4x upscaling ESRGAN FID 166.36 #9 of 11 Archive leaderboard report
Image Super-Resolution FFHQ 256 x 256 - 4x upscaling ESRGAN MS-SSIM 0.747 #9 of 11 Archive leaderboard report
Image Super-Resolution FFHQ 256 x 256 - 4x upscaling ESRGAN PSNR 15.43 #9 of 11 Archive leaderboard report
Image Super-Resolution FFHQ 256 x 256 - 4x upscaling ESRGAN SSIM 0.267 #9 of 11 Archive leaderboard report
Image Super-Resolution FFHQ 512 x 512 - 4x upscaling ESRGAN FED 0.1107 #7 of 8 Archive leaderboard report
Image Super-Resolution FFHQ 512 x 512 - 4x upscaling ESRGAN FID 3.503 #7 of 8 Archive leaderboard report
Image Super-Resolution FFHQ 512 x 512 - 4x upscaling ESRGAN LLE 2.261 #7 of 8 Archive leaderboard report
Image Super-Resolution FFHQ 512 x 512 - 4x upscaling ESRGAN LPIPS 0.1221 #7 of 8 Archive leaderboard report
Image Super-Resolution FFHQ 512 x 512 - 4x upscaling ESRGAN MS-SSIM 0.935 #7 of 8 Archive leaderboard report
Image Super-Resolution FFHQ 512 x 512 - 4x upscaling ESRGAN NIQE 6.984 #7 of 8 Archive leaderboard report
Image Super-Resolution FFHQ 512 x 512 - 4x upscaling ESRGAN PSNR 27.134 #7 of 8 Archive leaderboard report
Image Super-Resolution FFHQ 512 x 512 - 4x upscaling ESRGAN SSIM 0.741 #7 of 8 Archive leaderboard report
Image Super-Resolution Manga109 - 4x upscaling SRGAN + Residual-in-Residual Dense Block PSNR 31.66 #27 of 50 Archive leaderboard report
Image Super-Resolution Manga109 - 4x upscaling SRGAN + Residual-in-Residual Dense Block SSIM 0.9196 #27 of 50 Archive leaderboard report
Image Super-Resolution Manga109 - 4x upscaling bicubic PSNR 24.89 #47 of 50 Archive leaderboard report
Image Super-Resolution Manga109 - 4x upscaling bicubic SSIM 0.7866 #47 of 50 Archive leaderboard report
Image Super-Resolution PIRM-test ESRGAN NIQE 2.55 #2 of 3 Archive leaderboard report
Image Super-Resolution Set14 - 4x upscaling SRGAN + Residual-in-Residual Dense Block PSNR 28.99 #30 of 104 Archive leaderboard report
Image Super-Resolution Set14 - 4x upscaling SRGAN + Residual-in-Residual Dense Block SSIM 0.7917 #30 of 104 Archive leaderboard report
Image Super-Resolution Urban100 - 4x upscaling SRGAN + Residual-in-Residual Dense Block PSNR 27.03 #23 of 65 Archive leaderboard report
Image Super-Resolution Urban100 - 4x upscaling SRGAN + Residual-in-Residual Dense Block SSIM 0.8153 #23 of 65 Archive leaderboard report
Image Super-Resolution Urban100 - 4x upscaling bicubic PSNR 23.14 #62 of 65 Archive leaderboard report
Image Super-Resolution Urban100 - 4x upscaling bicubic SSIM 0.6577 #62 of 65 Archive leaderboard report
Video Super-Resolution MSU Video Super Resolution Benchmark: Detail Restoration ESRGAN 1 - LPIPS 0.948 #14 of 32 Archive leaderboard report
Video Super-Resolution MSU Video Super Resolution Benchmark: Detail Restoration ESRGAN ERQAv1.0 0.735 #14 of 32 Archive leaderboard report
Video Super-Resolution MSU Video Super Resolution Benchmark: Detail Restoration ESRGAN FPS 1.004 #14 of 32 Archive leaderboard report
Video Super-Resolution MSU Video Super Resolution Benchmark: Detail Restoration ESRGAN PSNR 27.33 #14 of 32 Archive leaderboard report
Video Super-Resolution MSU Video Super Resolution Benchmark: Detail Restoration ESRGAN QRCRv1.0 0 #14 of 32 Archive leaderboard report
Video Super-Resolution MSU Video Super Resolution Benchmark: Detail Restoration ESRGAN SSIM 0.808 #14 of 32 Archive leaderboard report
Video Super-Resolution MSU Video Super Resolution Benchmark: Detail Restoration ESRGAN Subjective score 5.353 #14 of 32 Archive leaderboard report
Video Super-Resolution MSU Video Upscalers: Quality Enhancement ESRGAN PSNR 27.29 #43 of 48 Archive leaderboard report
Video Super-Resolution MSU Video Upscalers: Quality Enhancement ESRGAN SSIM 0.936 #43 of 48 Archive leaderboard report
Video Super-Resolution MSU Video Upscalers: Quality Enhancement ESRGAN VMAF 56.69 #43 of 48 Archive leaderboard report

Ranks are positions in the archive's leaderboards as they stood at the 2025-07-28 snapshot. Results published since then are not among these rows, so a rank here is not a current standing.

Methods

Batch NormalizationConvolutionDense ConnectionsDropoutMax PoolingPReLUPixelShuffleReLURelativistic GANResidual BlockResidual ConnectionSRGANSRGAN Residual BlockSigmoid ActivationSoftmaxVGG Loss

Report a problem or propose a change · a person checks every report against the paper or source before anything changes; decisions are listed on /corrections