Papers › UniMatch V2: Pushing the Limit of Semi-Supervised Semantic Segmentation

UniMatch V2: Pushing the Limit of Semi-Supervised Semantic Segmentation

14 Oct 2024arXiv:2410.10777archive 2025-07-28

Lihe Yang, Zhen Zhao, Hengshuang Zhao

Semi-supervised semantic segmentation (SSS) aims at learning rich visual knowledge from cheap unlabeled images to enhance semantic segmentation capability. Among recent works, UniMatch improves its precedents tremendously by amplifying the practice of weak-to-strong consistency regularization. Subsequent works typically follow similar pipelines and propose various delicate designs. Despite the achieved progress, strangely, even in this flourishing era of numerous powerful vision models, almost all SSS works are still sticking to 1) using outdated ResNet encoders with small-scale ImageNet-1K pre-training, and 2) evaluation on simple Pascal and Cityscapes datasets. In this work, we argue that, it is necessary to switch the baseline of SSS from ResNet-based encoders to more capable ViT-based encoders (e.g., DINOv2) that are pre-trained on massive data. A simple update on the encoder (even using 2x fewer parameters) can bring more significant improvement than careful method designs. Built on this competitive baseline, we present our upgraded and simplified UniMatch V2, inheriting the core spirit of weak-to-strong consistency from V1, but requiring less training cost and providing consistently better results. Additionally, witnessing the gradually saturated performance on Pascal and Cityscapes, we appeal that we should focus on more challenging benchmarks with complex taxonomy, such as ADE20K and COCO datasets. Code, models, and logs of all reported values, are available at https://github.com/LiheYoung/UniMatch-V2.

PaperPDFCodeCode 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 5 code samples harvested from 1 repository linked to this paper: 5 ran (of which 0 constructed an object rather than computing a result; 2 with no instrument failure: 0 honoured, 0 violated, 2 with no contract checked; 3 where Syntology's instrument failed) · 0 unverified. Where Syntology's instrument failed: 1 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: 5 samples from 1 repository, 5 ran (2 with no instrument failure; 3 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 2026-09-24 and 2026-09-26 (UTC): the dates Syntology's graph wrote each sample's verification level, not when the code executed. 1 of the 5 samples that ran has no recorded date.

LiheYoung/UniMatch-V2 officialmentioned in papermentioned on GitHubpytorchMITSyntology: 5 samples, 5 ran (3 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

5 samples harvested: 5 ran (of which 0 constructed an object rather than computing a result; 2 with no instrument failure: 0 honoured, 0 violated, 2 with no contract checked; 3 where Syntology's instrument failed) · 0 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.

1ran · our draft was wrong
2ran; the contract check could not run
2ran

Licence: 3 of the 5 samples are pointer only, meaning Syntology does not serve that copy's text. For 3 of them, 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 LiheYoung/UniMatch-V2. “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.

LiheYoung/UniMatch-V2 official repository 5 samples, 5 ran (3 instrument-failed) · 0 unverified
crop in 99 papers
census: 5 implementations from 5 papers, 4 distinct outputs across 4 buckets, one shared input per bucket
dataset/transform.py ran MIT (permissive) here; the stored copy is pointer only · 21e12abf73350be0 · report fixture: signature img:Image/0/?, mask:Image/0/?, size:int/0/int, ignore_value:int/0/int · returned tuple · level recorded 2026-09-26
hflip in 29 papers dataset/transform.py ran; the contract check could not run (RAISES) MIT (permissive) here; the stored copy is pointer only · 9e96d1dddb1f71af · report fixture: signature img:Image/0/?, mask:Image/0/?, p:float/0/float · returned tuple · level recorded 2026-09-26
named_apply in 32 papers model/backbone/dinov2.py ran · our draft was wrong MIT (permissive) · e7fcb6d9ac9deaf4 · report fixture: signature fn:callable/0/?, module:callable/0/?, name:str/0/str, depth_first:bool/0/bool, include_root:bool/0/bool · returned SimpleModel · level recorded, date not recorded
normalize in 638 papers
census: 70 implementations from 82 papers, 51 distinct outputs across 11 buckets, one shared input per bucket
dataset/transform.py ran; the contract check could not run (RAISES) MIT (permissive) here; the stored copy is pointer only · 5717ec3a3f33f0a8 · report fixture: signature img:Image/0/?, mask:ndarray/2/int · returned tuple · level recorded 2026-09-26
setup_distributed in 11 papers util/dist_helper.py ran MIT (permissive) · 7c60f8e4d3bbd46f · report fixture: signature backend:str/0/str, port:int/0/int · returned tuple · level recorded 2026-09-24

Tasks

Semantic SegmentationSemi-Supervised Semantic SegmentationSemi-supervised Change Detection

Results from the paper archive 2025-07-28

TaskDatasetModelMetricValueRank at snapshotLeaderboardReport
Semi-Supervised Semantic Segmentation ADE20K 1/16 labeled UniMatch V2 Validation mIoU 46.7 #1 of 5 Archive leaderboard report
Semi-Supervised Semantic Segmentation ADE20K 1/32 labeled UniMatch V2 Validation mIoU 45.0 #1 of 5 Archive leaderboard report
Semi-Supervised Semantic Segmentation COCO 1/128 labeled UniMatch V2 Validation mIoU 58.7 #1 of 9 Archive leaderboard report
Semi-Supervised Semantic Segmentation COCO 1/256 labeled UniMatch V2 Validation mIoU 55.8 #1 of 9 Archive leaderboard report
Semi-Supervised Semantic Segmentation COCO 1/32 labeled UniMatch V2 Validation mIoU 63.3 #1 of 7 Archive leaderboard report
Semi-Supervised Semantic Segmentation COCO 1/512 labeled UniMatch V2 Validation mIoU 47.9 #2 of 8 Archive leaderboard report
Semi-Supervised Semantic Segmentation COCO 1/64 labeled UniMatch V2 Validation mIoU 60.4 #1 of 9 Archive leaderboard report
Semi-Supervised Semantic Segmentation Cityscapes 12.5% labeled UniMatch V2 (DINOv2-B) Validation mIoU 84.3% #1 of 33 Archive leaderboard report
Semi-Supervised Semantic Segmentation Cityscapes 25% labeled UniMatch V2 (DINOv2-B) Validation mIoU 84.5% #1 of 30 Archive leaderboard report
Semi-Supervised Semantic Segmentation Cityscapes 50% labeled UniMatch V2 (DINOv2-B) Validation mIoU 85.1% #1 of 23 Archive leaderboard report
Semi-Supervised Semantic Segmentation Cityscapes 6.25% labeled UniMatch V2 (DINOv2-B) Validation mIoU 83.6 #1 of 18 Archive leaderboard report
Semi-Supervised Semantic Segmentation PASCAL VOC 2012 1464 labels UniMatch V2 (DINOv2-B) Validation mIoU 90.8 #1 of 17 Archive leaderboard report
Semi-Supervised Semantic Segmentation PASCAL VOC 2012 183 labeled UniMatch V2 (DINOv2-B) Validation mIoU 87.9 #1 of 16 Archive leaderboard report
Semi-Supervised Semantic Segmentation PASCAL VOC 2012 366 labeled UniMatch V2 (DINOv2-B) Validation mIoU 88.9 #1 of 15 Archive leaderboard report
Semi-Supervised Semantic Segmentation PASCAL VOC 2012 732 labeled UniMatch V2 (DINOv2-B) Validation mIoU 90.0 #1 of 16 Archive leaderboard report
Semi-Supervised Semantic Segmentation PASCAL VOC 2012 92 labeled UniMatch V2 (DINOv2-B) Validation mIoU 86.3 #2 of 17 Archive leaderboard report
Semi-supervised Change Detection LEVIR-CD - 10% labeled data UniMatch V2 IoU 83.8 #1 of 5 Archive leaderboard report
Semi-supervised Change Detection LEVIR-CD - 10% labeled data UniMatch V2 OA 99.11 #1 of 5 Archive leaderboard report
Semi-supervised Change Detection LEVIR-CD - 20% labeled data UniMatch V2 IoU 84.3 #1 of 4 Archive leaderboard report
Semi-supervised Change Detection LEVIR-CD - 20% labeled data UniMatch V2 OA 99.14 #1 of 4 Archive leaderboard report
Semi-supervised Change Detection LEVIR-CD - 40% labeled data UniMatch V2 IoU 84.3 #2 of 4 Archive leaderboard report
Semi-supervised Change Detection LEVIR-CD - 40% labeled data UniMatch V2 OA 99.14 #2 of 4 Archive leaderboard report
Semi-supervised Change Detection LEVIR-CD - 5% labeled data UniMatch V2 IoU 83.3 #1 of 5 Archive leaderboard report
Semi-supervised Change Detection LEVIR-CD - 5% labeled data UniMatch V2 OA 99.08 #1 of 5 Archive leaderboard report
Semi-supervised Change Detection WHU - 20% labeled data UniMatch V2 IoU 87.9 #1 of 4 Archive leaderboard report
Semi-supervised Change Detection WHU - 20% labeled data UniMatch V2 OA 99.50 #1 of 4 Archive leaderboard report
Semi-supervised Change Detection WHU - 40% labeled data UniMatch V2 IoU 88.6 #1 of 4 Archive leaderboard report
Semi-supervised Change Detection WHU - 40% labeled data UniMatch V2 OA 99.52 #1 of 4 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

Average PoolingConvolutionFocusGlobal Average PoolingKaiming InitializationMax Pooling

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