Papers › Unsupervised Semantic Segmentation by Distilling Feature Correspondences

Unsupervised Semantic Segmentation by Distilling Feature Correspondences

16 Mar 2022ICLR 2022 4arXiv:2203.08414archive 2025-07-28

Mark Hamilton, Zhoutong Zhang, Bharath Hariharan, Noah Snavely, William T. Freeman

Unsupervised semantic segmentation aims to discover and localize semantically meaningful categories within image corpora without any form of annotation. To solve this task, algorithms must produce features for every pixel that are both semantically meaningful and compact enough to form distinct clusters. Unlike previous works which achieve this with a single end-to-end framework, we propose to separate feature learning from cluster compactification. Empirically, we show that current unsupervised feature learning frameworks already generate dense features whose correlations are semantically consistent. This observation motivates us to design STEGO (Self-supervised Transformer with Energy-based Graph Optimization), a novel framework that distills unsupervised features into high-quality discrete semantic labels. At the core of STEGO is a novel contrastive loss function that encourages features to form compact clusters while preserving their relationships across the corpora. STEGO yields a significant improvement over the prior state of the art, on both the CocoStuff (+14 mIoU) and Cityscapes (+9 mIoU) semantic segmentation challenges.

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 14 code samples harvested from 3 repositories linked to this paper: 11 ran (of which 0 constructed an object rather than computing a result; 7 with no instrument failure: 3 honoured, 1 violated, 3 with no contract checked; 4 where Syntology's instrument failed) · 3 unverified. Where Syntology's instrument failed: 2 where our contract draft was wrong; 2 where the contract check could not run; failures of Syntology's instrument, not of the code.

By repository: official repository: 3 samples from 1 repository, 3 ran (1 with no instrument failure; 2 where Syntology's instrument failed); community (archive-listed): 11 samples from 2 repositories, 8 ran (6 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.

Levels written between 2026-09-22 and 2026-09-24 (UTC): the dates Syntology's graph wrote each sample's verification level, not when the code executed. 2 of the 11 samples that ran have no recorded date.

mhamilton723/STEGO officialmentioned on GitHubpytorchSyntology: 3 samples, 3 ran (2 instrument-failed) · 0 unverified report
leggedrobotics/self_supervised_segmentation mentioned on GitHubpytorchMITSyntology: 10 samples, 7 ran (1 instrument-failed) · 3 unverified report
merantix-momentum/stego-studies mentioned on GitHubpytorchSyntology: 1 sample, 1 ran (1 instrument-failed) · 0 unverified 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

14 samples harvested: 11 ran (of which 0 constructed an object rather than computing a result; 7 with no instrument failure: 3 honoured, 1 violated, 3 with no contract checked; 4 where Syntology's instrument failed) · 3 unverified. “Where Syntology's instrument failed” counts failures of Syntology's instrument, not of the code. Syntology's record for this page last changed ; when this build read Syntology's graph is in the build record. That date is not when the samples ran.

3ran · honoured contract
1ran · violated contract
2ran · our draft was wrong
2ran; the contract check could not run
3ran
3unverified

Licence: 1 of the 14 samples is pointer only, meaning Syntology does not serve that copy's text. For that sample, this paper's copy is under a permissive licence and cleared, but the stored copy of the same code that Syntology serves is not; the licence label's tooltip gives Syntology's recorded reason. This page shows no code text for any sample; each one links to its file in the repository.

Harvested from 3 repositories linked to this paper, official or community; the list below is grouped by repository, each group headed by its name and what it is to the paper. “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.

mhamilton723/STEGO official repository 3 samples, 3 ran (2 instrument-failed) · 0 unverified
one_hot_feats src/train_segmentation.py ran · our draft was wrong MIT (permissive) · eb2b27a0c094426d · report fixture: signature labels:Tensor/3/int, n_classes:int/0/int · returned Tensor · level recorded 2026-09-22
resize in 108 papers
census: 10 implementations from 14 papers, 5 distinct outputs across 2 buckets, one shared input per bucket this sample: bucket 1 of 2, arg 1: rank 4, kind float, dtype float32
src/train_segmentation.py ran; the contract check could not run (RAISES) · metamorphic tier: invariant MIT (permissive) · 29fed21d5b05ba11 · report fixture: returned Tensor · level recorded 2026-09-22
shuffle in 35 papers src/train_segmentation.py ran · honoured contract MIT (permissive) · 205e834237bf794c · report fixture: returned Tensor · level recorded 2026-09-22
leggedrobotics/self_supervised_segmentation community (archive-listed) 10 samples, 7 ran (1 instrument-failed) · 3 unverified
bool_flag in 47 papers stego/backbones/dino/utils.py ran · violated contract MIT (permissive) · 4d25be8ebf17fde1 · report fixture: signature s:str/0/str · returned bool · level recorded 2026-09-24
clip_gradients in 31 papers stego/backbones/dino/utils.py ran · honoured contract MIT (permissive) · 2b5489053b222e00 · report fixture: signature model:callable/0/?, clip:float/0/float · returned list · level recorded 2026-09-24
cosine_scheduler in 74 papers stego/backbones/dino/utils.py ran · honoured contract fingerprinted MIT (permissive) · 361a6b24f11fc50a · report fixture: signature base_value:float/0/float, final_value:float/0/float, epochs:int/0/int, niter_per_ep:int/0/int, warmup_epochs:int/0/int, start_warmup_value:float/0/float · returned ndarray · level recorded, date not recorded
drop_path in 325 papers
census: 27 implementations from 61 papers, 5 distinct outputs across 2 buckets, one shared input per bucket this sample: bucket 2 of 2, arg 1: rank 2, kind float, dtype float32
stego/backbones/dino/vision_transformer.py ran; the contract check could not run (RAISES) · metamorphic tier: invariant fingerprinted MIT (permissive) here; the stored copy is pointer only · 55120f2026b56aa2 · report fixture: signature x:Tensor/2/float, drop_prob:float/0/float, training:bool/0/bool · returned Tensor · level recorded, date not recorded
get_nn_file_name stego/utils.py ran MIT (permissive) · a02d5b8c7dc6802d · report fixture: signature data_dir:str/0/str, dataset_name:str/0/str, model_type:str/0/str, image_set:str/0/str, resolution:str/0/str · returned str · level recorded 2026-09-24
get_transform in 75 papers stego/utils.py ran MIT (permissive) · 274d426926a54825 · report fixture: signature res:int/0/int, is_label:bool/0/bool, crop_type:str/0/str, is_tensor:bool/0/bool, do_normalize:bool/0/bool · returned Compose · level recorded 2026-09-24
load_image_to_tensor stego/utils.py ran MIT (permissive) · b1e60ab85b4a8299 · report fixture: signature img_path:str/0/str, resolution:int/0/int, brightness_factor:float/0/float, contrast_factor:float/0/float, saturation_factor:float/0/float, hue_factor:float/0/float, gaussian_sigma:NoneType/0/?, gaussian_kernel_size:NoneType/0/? · returned Tensor · level recorded 2026-09-24
get_backbone in 28 papers stego/backbones/backbone.py unverified MIT (permissive) · dbcfa76b17fae535 · report
vit_small in 33 papers stego/backbones/dino/vision_transformer.py unverified MIT (permissive) · afc44a0db78b7846 · report
vit_tiny in 40 papers stego/backbones/dino/vision_transformer.py unverified MIT (permissive) · a250f0f34719cf17 · report
merantix-momentum/stego-studies community (archive-listed) 1 sample, 1 ran (1 instrument-failed) · 0 unverified
get_class_labels in 6 papers src/train_segmentation.py ran · our draft was wrong MIT (permissive) · d6f15c1eec2a900c · report fixture: signature dataset_name:str/0/str · returned list · level recorded 2026-09-22

Tasks

FormSemantic SegmentationUnsupervised Semantic Segmentation

Results from the paper archive 2025-07-28

TaskDatasetModelMetricValueRank at snapshotLeaderboardReport
Unsupervised Semantic Segmentation COCO-Stuff-27 STEGO (ViT-B/8) Clustering [Accuracy] 56.9 #8 of 29 Archive leaderboard report
Unsupervised Semantic Segmentation COCO-Stuff-27 STEGO (ViT-B/8) Clustering [mIoU] 28.2 #8 of 29 Archive leaderboard report
Unsupervised Semantic Segmentation COCO-Stuff-27 STEGO (ViT-S/8) Clustering [mIoU] 24.5 #17 of 29 Archive leaderboard report
Unsupervised Semantic Segmentation COCO-Stuff-27 STEGO (ViT-S/8) Linear Classifier [Accuracy] 74.4 #17 of 29 Archive leaderboard report
Unsupervised Semantic Segmentation COCO-Stuff-27 STEGO (ViT-S/8) Linear Classifier [mIoU] 38.3 #17 of 29 Archive leaderboard report
Unsupervised Semantic Segmentation Cityscapes test STEGO Accuracy 73.2 #6 of 14 Archive leaderboard report
Unsupervised Semantic Segmentation Cityscapes test STEGO mIoU 21.0 #6 of 14 Archive leaderboard report
Unsupervised Semantic Segmentation Potsdam-3 STEGO Accuracy 77.0 #6 of 8 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

AdamAttentionCRFDINODropoutLayer NormalizationLinear LayerMulti-Head AttentionResidual ConnectionSoftmaxTransformerVision Transformer

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