Papers › Cost Aggregation with 4D Convolutional Swin Transformer for Few-Shot Segmentation

Cost Aggregation with 4D Convolutional Swin Transformer for Few-Shot Segmentation

22 Jul 2022arXiv:2207.10866archive 2025-07-28

Sunghwan Hong, Seokju Cho, Jisu Nam, Stephen Lin, Seungryong Kim

This paper presents a novel cost aggregation network, called Volumetric Aggregation with Transformers (VAT), for few-shot segmentation. The use of transformers can benefit correlation map aggregation through self-attention over a global receptive field. However, the tokenization of a correlation map for transformer processing can be detrimental, because the discontinuity at token boundaries reduces the local context available near the token edges and decreases inductive bias. To address this problem, we propose a 4D Convolutional Swin Transformer, where a high-dimensional Swin Transformer is preceded by a series of small-kernel convolutions that impart local context to all pixels and introduce convolutional inductive bias. We additionally boost aggregation performance by applying transformers within a pyramidal structure, where aggregation at a coarser level guides aggregation at a finer level. Noise in the transformer output is then filtered in the subsequent decoder with the help of the query's appearance embedding. With this model, a new state-of-the-art is set for all the standard benchmarks in few-shot segmentation. It is shown that VAT attains state-of-the-art performance for semantic correspondence as well, where cost aggregation also plays a central role.

PaperPDFCodeCode Syntology ranMethod as code

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 20 code samples: 18 from 1 repository linked to this paper, 2 identical to code harvested from another paper's repository. Across all 20: 12 ran (of which 7 constructed an object rather than computing a result; 8 with no instrument failure: 0 honoured, 0 violated, 8 with no contract checked; 4 where Syntology's instrument failed) · 8 unverified. Where Syntology's instrument failed: 3 where our contract draft was wrong; 1 where the contract check could not run; failures of Syntology's instrument, not of the code.

By repository: official repository: 18 samples from 1 repository, 10 ran (8 with no instrument failure; 2 where Syntology's instrument failed); 2 identical to code first harvested elsewhere. 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 2026-09-25 (UTC): the dates Syntology's graph wrote each sample's verification level, not when the code executed.

Seokju-Cho/Volumetric-Aggregation-Transformer officialmentioned on GitHubpytorchSyntology: 18 samples, 10 ran (2 instrument-failed) · 8 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

20 samples harvested: 12 ran (of which 7 constructed an object rather than computing a result; 8 with no instrument failure: 0 honoured, 0 violated, 8 with no contract checked; 4 where Syntology's instrument failed) · 8 unverified. “Where Syntology's instrument failed” counts failures 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.

3ran · our draft was wrong
1ran; the contract check could not run
8ran
8unverified

Licence: 2 of the 20 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 Seokju-Cho/Volumetric-Aggregation-Transformer. 2 samples are listed last, in a group without a repository. “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”. For 2 samples that ran, the graph records a signature this page does not show: for a class the fixture built without calling, the harness called the object with inputs it made up itself, so the recorded signature is the harness's probe, not the fixture's arguments. “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.

Seokju-Cho/Volumetric-Aggregation-Transformer official repository 18 samples, 10 ran (2 instrument-failed) · 8 unverified
Conv4d model/vat.py ran MIT (permissive) · a689dd29d10a5588 · report fixture: returned Conv4d · level recorded 2026-09-22
Correlation in 6 papers model/vat.py ran MIT (permissive) · d957466bd907d3e2 · report fixture: signature query_feats:list/1/Tensor, support_feats:list/1/Tensor, stack_ids:list/1/int · returned list · level recorded 2026-09-24
Interpolate4d model/vat.py ran MIT (permissive) · 97b275801cca78fc · report fixture: returned Interpolate4d · level recorded 2026-09-22
MaxPool4d model/vat.py ran MIT (permissive) · b966405e9e7c94fc · report fixture: returned MaxPool4d · level recorded 2026-09-22
PatchMerging in 31 papers model/vat.py ran MIT (permissive) · 688a7c109b6a8925 · report fixture: returned PatchMerging · level recorded 2026-09-22
TransformerWarper model/vat.py ran MIT (permissive) · 9f47dd1aa3379e7a · report fixture: returned TransformerWarper · level recorded 2026-09-22
TransformerWarper2d model/vat.py ran fingerprinted MIT (permissive) · 3c9c3d947cf636e4 · report fixture: returned TransformerWarper2d · level recorded 2026-09-22
WindowAttention in 50 papers model/vat.py ran MIT (permissive) · c45ba97aae362041 · report fixture: returned WindowAttention · level recorded 2026-09-22
extract_feat_res model/vat.py ran · our draft was wrong MIT (permissive) · 5e71dad5278d69a0 · report fixture: signature img:Tensor/4/float, backbone:callable/0/?, feat_ids:list/1/int, bottleneck_ids:list/1/int, lids:list/1/int · returned list · level recorded 2026-09-25
extract_feat_vgg model/vat.py ran · our draft was wrong MIT (permissive) · cda8eb3e776a0591 · report fixture: signature img:Tensor/4/float, backbone:callable/0/?, feat_ids:list/1/int, bottleneck_ids:NoneType/0/?, lids:NoneType/0/? · returned list · level recorded 2026-09-22
BasicLayer in 56 papers model/vat.py unverified MIT (permissive) · 30c94d4ce2dcc31b · report
Encoder4D model/vat.py unverified MIT (permissive) · f753e1094cc84a3f · report
OurModel model/vat.py unverified MIT (permissive) · fcf09c7f87136ef1 · report
SwinTransformer in 15 papers model/vat.py unverified MIT (permissive) · 211001493120f0e2 · report
SwinTransformer2d model/vat.py unverified MIT (permissive) · 4390e9116e6b7b05 · report
SwinTransformerBlock in 35 papers model/vat.py unverified MIT (permissive) · cf7fdf758691ee4a · report
VAT model/vat.py unverified MIT (permissive) · 09ecea11a241565f · report
VATLearner model/vat.py unverified MIT (permissive) · 616c158398e06b3e · report
Repository not located 2 samples, 2 ran (2 instrument-failed) · 0 unverified
2 samples are identical to code Syntology harvested from another paper's repository; this paper's own copy was not located, so no repository is named and no licence is recorded.
window_partition in 377 papers
census: 13 implementations from 17 papers, 5 distinct outputs across 3 buckets, one shared input per bucket
repository not located ran; the contract check could not run (RAISES) · metamorphic tier: invariant fingerprinted licence of this copy not recorded · f9fd6241d935f07b · report fixture: signature x:Tensor/4/float, window_size:int/0/int · returned Tensor · level recorded 2026-09-22
window_reverse in 296 papers repository not located ran · our draft was wrong licence of this copy not recorded · fb32094c6dbece71 · report fixture: signature windows:Tensor/4/float, window_size:int/0/int, H:int/0/int, W:int/0/int · returned Tensor · level recorded 2026-09-22

Method as code Syntology

What Syntology's graph records as this paper's claimed method, lifted into code: each assembly names its bricks (functions and classes), each with its own role and its own verification level. An assembly is not a check that the code implements the method correctly or reproduces the paper's results. Levels are each brick's own verification level in Syntology's graph (level 0: no recorded run), listed brick by brick and never combined into one level; “ran” means the brick executed on a synthesized input, not that its output is correct, and the instrument figure counts failures of Syntology's instrument, not of the code. Where a model read the code against the paper's claim, each assembly's label verdict is one model's judgement, kept in this page's JSON twin; it is not calibrated, so no card prints it.

1 assembly on this paper: 1 harvested, 0 generated.

harvested Assembly for Volumetric Aggregation with Transformers (VAT) (graph method: Volumetric Aggregation with Transformers (VAT)). 21 bricks; individually on a fixture: 13 ran (of which 8 constructed an object rather than computing a result; 9 with no instrument failure: 0 honoured, 0 violated, 9 with no contract checked; 4 where Syntology's instrument failed) · 8 unverified. The assembly has not been run end-to-end.

  1. VAT entry · level 0 · unverified · 09ecea11a241565f
  2. Correlation helper · level 1 · ran · d957466bd907d3e2
  3. VATLearner helper · level 0 · unverified · 616c158398e06b3e
  4. extract_feat_res helper · level 1 · ran · our draft was wrong · 5e71dad5278d69a0
  5. extract_feat_vgg helper · level 1 · ran · our draft was wrong · cda8eb3e776a0591
  6. OurModel helper · level 0 · unverified · fcf09c7f87136ef1
  7. Encoder4D model · level 0 · unverified · f753e1094cc84a3f
  8. Interpolate4d helper · level 1 · ran · 97b275801cca78fc
  9. SwinTransformer helper · level 0 · unverified · 211001493120f0e2
  10. SwinTransformer2d helper · level 0 · unverified · 4390e9116e6b7b05
  11. TransformerWarper model · level 1 · ran · 9f47dd1aa3379e7a
  12. TransformerWarper2d helper · level 1 · ran · 3c9c3d947cf636e4
  13. Conv4d model · level 1 · ran · a689dd29d10a5588
  14. BasicLayer model · level 0 · unverified · 30c94d4ce2dcc31b
  15. PatchMerging helper · level 1 · ran · 688a7c109b6a8925
  16. MaxPool4d helper · level 1 · ran · b966405e9e7c94fc
  17. SwinTransformerBlock model · level 0 · unverified · cf7fdf758691ee4a
  18. Mlp helper · level 1 · ran · a1715055832f71fd
  19. WindowAttention helper · level 1 · ran · c45ba97aae362041
  20. window_partition helper · level 1 · ran; the contract check could not run (RAISES) · f9fd6241d935f07b
  21. window_reverse helper · level 1 · ran · our draft was wrong · fb32094c6dbece71

Each brick ends with its code_sha256, Syntology's identity for that exact code; click one to copy the get_code call that returns it. For agents, on Syntology's MCP service: assembly(arxiv_id="2207.10866") returns these assemblies with each brick's role and wiring (how to connect).

Tasks

DecoderFew-Shot Semantic SegmentationInductive BiasSemantic correspondence

Results from the paper archive 2025-07-28

TaskDatasetModelMetricValueRank at snapshotLeaderboardReport
Few-Shot Semantic Segmentation COCO-20i (1-shot) VAT (ResNet-101) FB-IoU 68.8 #58 of 85 Archive leaderboard report
Few-Shot Semantic Segmentation COCO-20i (1-shot) VAT (ResNet-101) Mean IoU 41.3 #58 of 85 Archive leaderboard report
Few-Shot Semantic Segmentation COCO-20i (5-shot) VAT (ResNet-101) FB-IoU 72.4 #51 of 81 Archive leaderboard report
Few-Shot Semantic Segmentation COCO-20i (5-shot) VAT (ResNet-101) Mean IoU 47.9 #51 of 81 Archive leaderboard report
Few-Shot Semantic Segmentation FSS-1000 (1-shot) VAT (ResNet-101) FB-IoU 94 #4 of 24 Archive leaderboard report
Few-Shot Semantic Segmentation FSS-1000 (1-shot) VAT (ResNet-101) Mean IoU 90.3 #4 of 24 Archive leaderboard report
Few-Shot Semantic Segmentation FSS-1000 (1-shot) VAT (ResNet-50) FB-IoU 93.8 #6 of 24 Archive leaderboard report
Few-Shot Semantic Segmentation FSS-1000 (1-shot) VAT (ResNet-50) Mean IoU 90.1 #6 of 24 Archive leaderboard report
Few-Shot Semantic Segmentation FSS-1000 (5-shot) VAT (ResNet-101) FB-IoU 94.4 #3 of 22 Archive leaderboard report
Few-Shot Semantic Segmentation FSS-1000 (5-shot) VAT (ResNet-101) Mean IoU 90.8 #3 of 22 Archive leaderboard report
Few-Shot Semantic Segmentation FSS-1000 (5-shot) VAT (ResNet-50) FB-IoU 94.2 #4 of 22 Archive leaderboard report
Few-Shot Semantic Segmentation FSS-1000 (5-shot) VAT (ResNet-50) Mean IoU 90.7 #4 of 22 Archive leaderboard report
Few-Shot Semantic Segmentation PASCAL-5i (1-Shot) VAT (ResNet-101) FB-IoU 79.6 #21 of 105 Archive leaderboard report
Few-Shot Semantic Segmentation PASCAL-5i (1-Shot) VAT (ResNet-101) Mean IoU 67.9 #21 of 105 Archive leaderboard report
Few-Shot Semantic Segmentation PASCAL-5i (1-Shot) VAT (ResNet-50) FB-IoU 77.8 #48 of 105 Archive leaderboard report
Few-Shot Semantic Segmentation PASCAL-5i (1-Shot) VAT (ResNet-50) Mean IoU 65.5 #48 of 105 Archive leaderboard report
Few-Shot Semantic Segmentation PASCAL-5i (5-Shot) VAT (ResNet-101) FB-IoU 83.2 #19 of 96 Archive leaderboard report
Few-Shot Semantic Segmentation PASCAL-5i (5-Shot) VAT (ResNet-101) Mean IoU 72 #19 of 96 Archive leaderboard report
Few-Shot Semantic Segmentation PASCAL-5i (5-Shot) VAT (ResNet-50) FB-IoU 80.9 #41 of 96 Archive leaderboard report
Few-Shot Semantic Segmentation PASCAL-5i (5-Shot) VAT (ResNet-50) Mean IoU 70.1 #41 of 96 Archive leaderboard report
Semantic correspondence PF-PASCAL VAT (ECCV) PCK 92.3 #8 of 15 Archive leaderboard report
Semantic correspondence PF-WILLOW VAT (ECCV) PCK 81.6 #2 of 8 Archive leaderboard report
Semantic correspondence SPair-71k VAT (ECCV) PCK 55.5 #12 of 22 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

Absolute Position EncodingsAdamAttentionBPEDense ConnectionsDropoutLabel SmoothingLayer NormalizationLinear LayerMulti-Head AttentionPosition-Wise Feed-Forward LayerResidual ConnectionSoftmaxStochastic DepthSwin TransformerTransformer

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