Papers › The Unreasonable Effectiveness of Deep Features as a Perceptual Metric

The Unreasonable Effectiveness of Deep Features as a Perceptual Metric

11 Jan 2018CVPR 2018 6arXiv:1801.03924archive 2025-07-28

Richard Zhang, Phillip Isola, Alexei A. Efros, Eli Shechtman, Oliver Wang

While it is nearly effortless for humans to quickly assess the perceptual similarity between two images, the underlying processes are thought to be quite complex. Despite this, the most widely used perceptual metrics today, such as PSNR and SSIM, are simple, shallow functions, and fail to account for many nuances of human perception. Recently, the deep learning community has found that features of the VGG network trained on ImageNet classification has been remarkably useful as a training loss for image synthesis. But how perceptual are these so-called "perceptual losses"? What elements are critical for their success? To answer these questions, we introduce a new dataset of human perceptual similarity judgments. We systematically evaluate deep features across different architectures and tasks and compare them with classic metrics. We find that deep features outperform all previous metrics by large margins on our dataset. More surprisingly, this result is not restricted to ImageNet-trained VGG features, but holds across different deep architectures and levels of supervision (supervised, self-supervised, or even unsupervised). Our results suggest that perceptual similarity is an emergent property shared across deep visual representations.

PaperPDFConference PDFCodeCode Syntology ran

In Syntology View this paper on Syntology, its page in Syntology's graph. That page adds which of its functions are also present in other papers' code Syntology holds, with dates, and its dated run record as JSON and a README badge. It also lists the repositories linked to the paper, the abstract and the calls for agents.

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

For agents, on Syntology's MCP service (how to connect):

Code

Syntology 2 code samples harvested from 1 repository linked to this paper: 2 ran (of which 0 constructed an object rather than computing a result; 0 with no instrument failure: 0 honoured, 0 violated, 0 with no contract checked; 2 where Syntology's instrument failed) · 0 unverified. Where Syntology's instrument failed: 1 where our contract draft was wrong; 1 where the contract check could not run; failures of Syntology's instrument, not of the code. Every run on this page is a failure of Syntology's instrument, not of the code.

By repository: official repository: 2 samples from 1 repository, 2 ran (0 with no instrument failure; 2 where Syntology's instrument failed). The run record, sample by sample. “Ran” means executed on a synthesized input, not that the code is correct or reproduces the paper.

No sample is linked to this paper from 3 of its repositories that Syntology harvested for other papers:

Levels written on 2026-09-22 (UTC): the dates Syntology's graph wrote each sample's verification level, not when the code executed.

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

richzhang/PerceptualSimilarity officialmentioned in papermentioned on GitHubpytorchSyntology: 2 samples, 2 ran (2 instrument-failed) · 0 unverified report
EndyWon/Deep-Feature-Perturbation mentioned on GitHubpytorchMITSyntology: not harvested report
Image-X-Institute/lpips_torch2tf mentioned on GitHubpytorchSyntology: not harvested report
Meghraj-Webllisto/stylegan mentioned on GitHubtfNOASSERTIONSyntology: not harvested report
Mohanned-Elkholy/BigGan-model-inversion-with-image-transformations mentioned on GitHubpytorchMITSyntology: no sample linked to this paper (harvested for another paper). Syntology's graph links these samples to On the "steerability" of generative adversarial networks (arXiv:1907.07171; that paper's own run record: 4 ran (of which 0 constructed an object rather than computing a result; 4 with no instrument failure: 0 honoured, 0 violated, 4 with no contract checked; 0 where Syntology's instrument failed) · 1 unverified); nothing checks that they implement this paper's method. report
Puzer/stylegan-encoder mentioned on GitHubtfSyntology: not harvested report
RudreshVeerkhare/StyleGan mentioned on GitHubtfSyntology: not harvested report
SUPERSHOPxyz/stylegan3-gradient mentioned on GitHubpytorchNOASSERTIONSyntology: not harvested report
ariel415el/PerceptualLossGLO-Pytorch mentioned on GitHubpytorchSyntology: not harvested report
ashutosh1919/FaceGenerationStyleGAN mentioned on GitHubtfNOASSERTIONSyntology: not harvested report
ayushgupta9198/gan mentioned on GitHubtfNOASSERTIONSyntology: not harvested report
ayushgupta9198/stylegan mentioned on GitHubtfNOASSERTIONSyntology: not harvested report
bytedance/LatentSync mentioned on GitHubpytorchSyntology: no sample linked to this paper (harvested for another paper). Syntology's graph links these samples to LatentSync: Audio Conditioned Latent Diffusion Models for Lip Sync (arXiv:2412.09262; that paper's own run record: 1 ran (of which 0 constructed an object rather than computing a result; 0 with no instrument failure: 0 honoured, 0 violated, 0 with no contract checked; 1 where Syntology's instrument failed) · 0 unverified); nothing checks that they implement this paper's method. report
cassava-math-ubb/experiments mentioned on GitHubtfSyntology: not harvested report
isaacschaal/SG_training mentioned on GitHubtfSyntology: not harvested report
jooyae/NVIDIA_STYLEGAN mentioned on GitHubtfNOASSERTIONSyntology: not harvested report
khurram702/StyleBasedGAN mentioned on GitHubtfNOASSERTIONSyntology: not harvested report
kozistr/gan-metrics mentioned on GitHubpytorchApache-2.0Syntology: no sample linked to this paper (harvested for another paper). Syntology's graph links these samples to Reliable Fidelity and Diversity Metrics for Generative Models (arXiv:2002.09797; that paper's own run record: 5 ran (of which 0 constructed an object rather than computing a result; 3 with no instrument failure: 1 honoured, 0 violated, 2 with no contract checked; 2 where Syntology's instrument failed) · 4 unverified); nothing checks that they implement this paper's method. report
pbaylies/stylegan-encoder mentioned on GitHubtfSyntology: not harvested report
ryanxingql/image-quality-assessment-toolbox mentioned on GitHubpytorchSyntology: not harvested report
stefkim/stylegan-batik mentioned on GitHubtfNOASSERTIONSyntology: not harvested report
woctezuma/stylegan2-projecting-images mentioned on GitHubtfMITSyntology: not harvested report
zzz2010/starganv2_paddle mentioned on GitHubpytorchNOASSERTIONSyntology: not harvested report
S-aiueo32/lpips-pytorch pytorchSyntology: not harvested 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.

A paper named beside a repository with no sample linked to this paper is shown with that paper's own run record, not this paper's: “ran” means executed on a synthesized input, not that the code is correct, and “where Syntology's instrument failed” counts failures of Syntology's instrument, not of the code. Papers are listed in arXiv-id order, at most three per repository.

Code Syntology ran Syntology

2 samples harvested: 2 ran (of which 0 constructed an object rather than computing a result; 0 with no instrument failure: 0 honoured, 0 violated, 0 with no contract checked; 2 where Syntology's instrument failed) · 0 unverified. “Where Syntology's instrument failed” counts failures of Syntology's instrument, not of the code. Every run on this page is a failure of Syntology's instrument, not of the code. Syntology's record for this page has not changed since , the first build that kept a record date for it; when this build read Syntology's graph is in the build record. That date is not when the samples ran.

1ran · our draft was wrong
1ran; the contract check could not run

Licence: 0 of the 2 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 richzhang/PerceptualSimilarity. “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. A line beginning “fixture:” under a sample that ran gives the arguments Syntology's fixture called the sample with, each as name:type/rank/element kind (? where none was recorded), and the type of what the fixture's run returned, exactly as the harness recorded them; the arguments and the returned type were recorded by two separate runs of the same fixture. No output value is shown, and a returned type says nothing about whether the output was right. “Level recorded” gives the UTC date when Syntology's graph wrote the sample's verification level, not when the code executed; where the level was written before Syntology's graph kept that date, the row says “level recorded, date not recorded”. “Honoured” and “violated” refer to a contract Syntology drafted from the code itself. “Our draft was wrong” and “the contract check could not run” are failures of Syntology's instrument, not of the code: the sample ran on Syntology's synthesized input, and then the drafted contract did not describe it, or the check itself raised an error (RAISES), lacked a dependency (DEP_MISSING) or ran out of time (TIMEOUT); hover the label for which. A metamorphic tier, where shown, is evidence of the output's shape, not that the sample implements the method.

Each sample ends with its code_sha256. A sample's code_sha256 is Syntology's identity for that exact code. It is printed in full (16 hexadecimal characters for every sample in this build), and get_code(code_sha256="…") takes it exactly as printed. An agent calls it on Syntology's MCP service (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 are grouped by repository: a repository marked official first, then by how directly the repository is tied to the paper, then in the archive's order; never by how many ran. 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 six 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; the licence is outside the permissive list Syntology serves text under (MIT, Apache-2.0, BSD and similar); this paper's copy was not located, so its licence is not recorded; or this paper's copy is cleared, but the stored copy of the same code that Syntology serves is not. 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.

A function or class name links to its page when the name occurs in the harvested code of at least 5 papers; the name is a string, not an identity. A census tag appears when Syntology's census ran implementations with that name on one shared input per bucket: it gives how many were compared, from how many papers, and how many distinct outputs they produced, and names this sample's bucket when this sample was one of them.

richzhang/PerceptualSimilarity official repository 2 samples, 2 ran (2 instrument-failed) · 0 unverified
spatial_average in 34 papers lpips/lpips.py ran · our draft was wrong fingerprinted BSD-2-Clause (permissive) · 29e5e72bcd006dcd · report fixture: signature in_tens:Tensor/4/float, keepdim:bool/0/bool · returned Tensor · level recorded 2026-09-22
upsample in 87 papers
census: 11 implementations from 16 papers, 9 distinct outputs across 3 buckets, one shared input per bucket this sample: bucket 1 of 3, arg 1: rank 4, kind float, dtype float32
lpips/lpips.py ran; the contract check could not run (RAISES) · metamorphic tier: invariant BSD-2-Clause (permissive) · 8cafd18b57a4d171 · report fixture: signature in_tens:Tensor/4/float, out_HW:tuple/1/int · returned Tensor · level recorded 2026-09-22

Tasks

Image Quality AssessmentSSIMVideo Quality Assessment

Datasets

Introduced by this paper, per the archive.

Perceptual Similarity

Results from the paper archive 2025-07-28

TaskDatasetModelMetricValueRank at snapshotLeaderboardReport
Video Quality Assessment MSU FR VQA Database LPIPS KLCC 0.5846 #19 of 20 Archive leaderboard report
Video Quality Assessment MSU FR VQA Database LPIPS PLCC 0.8128 #19 of 20 Archive leaderboard report
Video Quality Assessment MSU FR VQA Database LPIPS SRCC 0.7538 #19 of 20 Archive leaderboard report
Video Quality Assessment MSU SR-QA Dataset LPIPS (Alex) KLCC 0.43158 #28 of 60 Archive leaderboard report
Video Quality Assessment MSU SR-QA Dataset LPIPS (Alex) PLCC 0.52385 #28 of 60 Archive leaderboard report
Video Quality Assessment MSU SR-QA Dataset LPIPS (Alex) SROCC 0.54461 #28 of 60 Archive leaderboard report
Video Quality Assessment MSU SR-QA Dataset LPIPS (Alex) Type FR #28 of 60 Archive leaderboard report
Video Quality Assessment MSU SR-QA Dataset LPIPS (VGG) KLCC 0.41471 #32 of 60 Archive leaderboard report
Video Quality Assessment MSU SR-QA Dataset LPIPS (VGG) PLCC 0.52820 #32 of 60 Archive leaderboard report
Video Quality Assessment MSU SR-QA Dataset LPIPS (VGG) SROCC 0.52868 #32 of 60 Archive leaderboard report
Video Quality Assessment MSU SR-QA Dataset LPIPS (VGG) Type FR #32 of 60 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

ConvolutionDense ConnectionsDropoutMax PoolingReLUSoftmax

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