Papers › Rethinking BiSeNet For Real-time Semantic Segmentation

Rethinking BiSeNet For Real-time Semantic Segmentation

27 Apr 2021CVPR 2021 1arXiv:2104.13188archive 2025-07-28

Mingyuan Fan, Shenqi Lai, Junshi Huang, Xiaoming Wei, Zhenhua Chai, Junfeng Luo, Xiaolin Wei

BiSeNet has been proved to be a popular two-stream network for real-time segmentation. However, its principle of adding an extra path to encode spatial information is time-consuming, and the backbones borrowed from pretrained tasks, e.g., image classification, may be inefficient for image segmentation due to the deficiency of task-specific design. To handle these problems, we propose a novel and efficient structure named Short-Term Dense Concatenate network (STDC network) by removing structure redundancy. Specifically, we gradually reduce the dimension of feature maps and use the aggregation of them for image representation, which forms the basic module of STDC network. In the decoder, we propose a Detail Aggregation module by integrating the learning of spatial information into low-level layers in single-stream manner. Finally, the low-level features and deep features are fused to predict the final segmentation results. Extensive experiments on Cityscapes and CamVid dataset demonstrate the effectiveness of our method by achieving promising trade-off between segmentation accuracy and inference speed. On Cityscapes, we achieve 71.9% mIoU on the test set with a speed of 250.4 FPS on NVIDIA GTX 1080Ti, which is 45.2% faster than the latest methods, and achieve 76.8% mIoU with 97.0 FPS while inferring on higher resolution images.

PaperPDFConference PDFCodeCode 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 16 code samples harvested from 4 repositories linked to this paper: 9 ran (of which 0 constructed an object rather than computing a result; 8 with no instrument failure: 0 honoured, 0 violated, 8 with no contract checked; 1 where Syntology's instrument failed) · 7 unverified. Where Syntology's instrument failed: 1 where the contract check could not run; failures of Syntology's instrument, not of the code.

By repository: official repository: 4 samples from 1 repository, 2 ran (2 with no instrument failure; 0 where Syntology's instrument failed); community (archive-listed): 12 samples from 3 repositories, 7 ran (6 with no instrument failure; 1 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 1 of its repositories that Syntology harvested for other papers:

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.

MichaelFan01/STDC-Seg officialmentioned in papermentioned on GitHubpytorchSyntology: 4 samples, 2 ran (0 instrument-failed) · 2 unverified report
Deci-AI/super-gradients mentioned on GitHubpytorchSyntology: 5 samples, 2 ran (0 instrument-failed) · 3 unverified report
pideyi1025/DeepLabV3Plus-RailSem19 mentioned on GitHubpytorchMITSyntology: 2 samples, 2 ran (1 instrument-failed) · 0 unverified report
zh320/realtime-semantic-segmentation-pytorch mentioned on GitHubpytorchSyntology: 5 samples, 3 ran (0 instrument-failed) · 2 unverified report
CuberrChen/STDCNet-Paddle paddleSyntology: not harvested report
PaddlePaddle/PaddleSeg paddleSyntology: no sample linked to this paper (harvested for another paper). Syntology's graph links these samples to EdgeFlow: Achieving Practical Interactive Segmentation with Edge-Guided Flow (arXiv:2109.09406; that paper's own run record: 5 ran (of which 0 constructed an object rather than computing a result; 3 with no instrument failure: 0 honoured, 0 violated, 3 with no contract checked; 2 where Syntology's instrument failed) · 0 unverified); nothing checks that they implement this paper's method. 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

16 samples harvested: 9 ran (of which 0 constructed an object rather than computing a result; 8 with no instrument failure: 0 honoured, 0 violated, 8 with no contract checked; 1 where Syntology's instrument failed) · 7 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; the contract check could not run
8ran
7unverified

Licence: 1 of the 16 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 4 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.

MichaelFan01/STDC-Seg official repository 4 samples, 2 ran (0 instrument-failed) · 2 unverified
CatBottleneck nets/stdcnet.py ran fingerprinted MIT (permissive) · 97c83f5a9b50d09c · report fixture: signature x:Tensor/4/float · returned Tensor · level recorded 2026-09-24
ConvX nets/stdcnet.py ran · metamorphic tier: invariant fingerprinted MIT (permissive) · bd9e2f16a5c9b881 · report fixture: signature x:Tensor/4/float · returned Tensor · level recorded 2026-09-22
AddBottleneck nets/stdcnet.py unverified MIT (permissive) · e352a650699c3703 · report
STDCNet1446 nets/stdcnet.py unverified MIT (permissive) · 6d18c795468569a2 · report
Deci-AI/super-gradients community (archive-listed) 5 samples, 2 ran (0 instrument-failed) · 3 unverified
ConvBNAct in 5 papers src/super_gradients/training/models/segmentation_models/stdc.py ran fingerprinted Apache-2.0 (permissive) · 0a4552c5e0d129a9 · report fixture: signature x:Tensor/4/float · returned Tensor · level recorded 2026-09-24
Residual in 24 papers src/super_gradients/training/models/segmentation_models/stdc.py ran · metamorphic tier: invariant fingerprinted Apache-2.0 (permissive) · 9593929516a67196 · report fixture: signature x:Tensor/2/float · returned Tensor · level recorded 2026-09-22
ConvBNReLU in 16 papers src/super_gradients/training/models/segmentation_models/stdc.py unverified Apache-2.0 (permissive) · 8942f44b7273d2b1 · report
STDCBlock src/super_gradients/training/models/segmentation_models/stdc.py unverified Apache-2.0 (permissive) · be030794ada81ba9 · report
SupportsReplaceInputChannels src/super_gradients/training/models/segmentation_models/stdc.py unverified Apache-2.0 (permissive) · 9669048dacdf270a · report
pideyi1025/DeepLabV3Plus-RailSem19 community (archive-listed) 2 samples, 2 ran (1 instrument-failed) · 0 unverified
get_preprocessing evaluate.py ran MIT (permissive) · 30208284736b9b4e · report fixture: signature preprocessing_fn:callable/0/? · returned Compose · level recorded 2026-09-24
to_tensor in 92 papers
census: 8 implementations from 8 papers, 3 distinct outputs across 3 buckets, one shared input per bucket
evaluate.py ran; the contract check could not run (RAISES) fingerprinted MIT (permissive) here; the stored copy is pointer only · 233ef2824a5baa1a · report fixture: signature x:ndarray/3/float · returned ndarray · level recorded 2026-09-24
zh320/realtime-semantic-segmentation-pytorch community (archive-listed) 5 samples, 3 ran (0 instrument-failed) · 2 unverified
AttentionRefinementModule models/stdc.py ran · metamorphic tier: invariant fingerprinted Apache-2.0 (permissive) · 651a44aa15b71684 · report fixture: signature x:Tensor/4/float · returned Tensor · level recorded 2026-09-22
FeatureFusionModule models/stdc.py ran · metamorphic tier: invariant Apache-2.0 (permissive) · 8ccfe4aa060066bc · report fixture: signature x_low:Tensor/4/float, x_high:Tensor/4/float · returned Tensor · level recorded 2026-09-22
STDCModule models/stdc.py ran · metamorphic tier: invariant fingerprinted Apache-2.0 (permissive) · a6375542a388edf0 · report fixture: signature x:Tensor/4/float · returned Tensor · level recorded 2026-09-22
STDC models/stdc.py unverified Apache-2.0 (permissive) · 09fe1398d60fd154 · report
SegHead models/stdc.py unverified Apache-2.0 (permissive) · 548ed9ee6767dd7a · report

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. Siblings are assemblies lifted from this paper in the same batch for the same method: alternative readings kept side by side; none replaces or supersedes another.

3 assemblies on this paper: 3 harvested, 0 generated, listed in 1 groups (siblings lifted in one batch form one group).

3 sibling assemblies for Short-Term Dense Concatenate network (STDC network) (harvested), lifted in the same batch on 2026-09-09: alternative readings kept side by side; none replaces or supersedes another.

harvested Assembly for Short-Term Dense Concatenate network (STDC network) (graph method: Short-Term Dense Concatenate network (STDC network)). 9 bricks; individually on a fixture: 6 ran (of which 1 constructed an object rather than computing a result; 5 with no instrument failure: 0 honoured, 0 violated, 5 with no contract checked; 1 where Syntology's instrument failed) · 3 unverified. The assembly has not been run end-to-end.

one of 3 siblings

  1. STDC entry · level 0 · unverified · 09fe1398d60fd154
  2. AttentionRefinementModule helper · level 1 · ran · 651a44aa15b71684
  3. ConvBNAct helper · level 1 · ran · baac3ba72d9b0da1
  4. FeatureFusionModule helper · level 1 · ran · 8ccfe4aa060066bc
  5. STDCModule model · level 1 · ran · a6375542a388edf0
  6. SegHead helper · level 0 · unverified · 548ed9ee6767dd7a
  7. conv1x1 helper · level 1 · ran · our draft was wrong · 1cb17a8a8a22bc6c
  8. register_model helper · level 0 · unverified · 70f1b7f8523ba6a8
  9. Activation helper · level 1 · ran · f3dd5914b236afcb

harvested Assembly for Short-Term Dense Concatenate network (STDC network) (graph method: Short-Term Dense Concatenate network (STDC network)). 5 bricks; individually on a fixture: 2 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; 0 where Syntology's instrument failed) · 3 unverified. The assembly has not been run end-to-end.

one of 3 siblings

  1. STDCBlock entry · level 0 · unverified · be030794ada81ba9
  2. ConvBNReLU helper · level 0 · unverified · 8942f44b7273d2b1
  3. Residual helper · level 1 · ran · 9593929516a67196
  4. ConvBNAct helper · level 1 · ran · 0a4552c5e0d129a9
  5. SupportsReplaceInputChannels helper · level 0 · unverified · 9669048dacdf270a

harvested Assembly for Short-Term Dense Concatenate network (STDC network) (graph method: Short-Term Dense Concatenate network (STDC network)). 4 bricks; individually on a fixture: 2 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; 0 where Syntology's instrument failed) · 2 unverified. The assembly has not been run end-to-end.

one of 3 siblings

  1. STDCNet1446 entry · level 0 · unverified · 6d18c795468569a2
  2. AddBottleneck model · level 0 · unverified · e352a650699c3703
  3. CatBottleneck model · level 1 · ran · 97c83f5a9b50d09c
  4. ConvX helper · level 1 · ran · bd9e2f16a5c9b881

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="2104.13188") returns these assemblies with each brick's role and wiring (how to connect).

Tasks

DecoderDichotomous Image SegmentationImage ClassificationImage SegmentationReal-Time Semantic SegmentationSegmentationSemantic Segmentationimage-classification

Results from the paper archive 2025-07-28

TaskDatasetModelMetricValueRank at snapshotLeaderboardReport
Dichotomous Image Segmentation DIS-TE1 STDC E-measure 0.798 #12 of 22 Archive leaderboard report
Dichotomous Image Segmentation DIS-TE1 STDC HCE 249 #12 of 22 Archive leaderboard report
Dichotomous Image Segmentation DIS-TE1 STDC MAE 0.090 #12 of 22 Archive leaderboard report
Dichotomous Image Segmentation DIS-TE1 STDC S-Measure 0.723 #12 of 22 Archive leaderboard report
Dichotomous Image Segmentation DIS-TE1 STDC max F-Measure 0.648 #12 of 22 Archive leaderboard report
Dichotomous Image Segmentation DIS-TE1 STDC weighted F-measure 0.562 #12 of 22 Archive leaderboard report
Dichotomous Image Segmentation DIS-TE2 STDC E-measure 0.834 #13 of 22 Archive leaderboard report
Dichotomous Image Segmentation DIS-TE2 STDC HCE 556 #13 of 22 Archive leaderboard report
Dichotomous Image Segmentation DIS-TE2 STDC MAE 0.092 #13 of 22 Archive leaderboard report
Dichotomous Image Segmentation DIS-TE2 STDC S-Measure 0.759 #13 of 22 Archive leaderboard report
Dichotomous Image Segmentation DIS-TE2 STDC max F-Measure 0.720 #13 of 22 Archive leaderboard report
Dichotomous Image Segmentation DIS-TE2 STDC weighted F-measure 0.636 #13 of 22 Archive leaderboard report
Dichotomous Image Segmentation DIS-TE3 STDC E-measure 0.855 #16 of 22 Archive leaderboard report
Dichotomous Image Segmentation DIS-TE3 STDC HCE 1081 #16 of 22 Archive leaderboard report
Dichotomous Image Segmentation DIS-TE3 STDC MAE 0.090 #16 of 22 Archive leaderboard report
Dichotomous Image Segmentation DIS-TE3 STDC S-Measure 0.771 #16 of 22 Archive leaderboard report
Dichotomous Image Segmentation DIS-TE3 STDC max F-Measure 0.745 #16 of 22 Archive leaderboard report
Dichotomous Image Segmentation DIS-TE3 STDC weighted F-measure 0.662 #16 of 22 Archive leaderboard report
Dichotomous Image Segmentation DIS-TE4 STDC E-measure 0.841 #14 of 22 Archive leaderboard report
Dichotomous Image Segmentation DIS-TE4 STDC HCE 3819 #14 of 22 Archive leaderboard report
Dichotomous Image Segmentation DIS-TE4 STDC MAE 0.102 #14 of 22 Archive leaderboard report
Dichotomous Image Segmentation DIS-TE4 STDC S-Measure 0.762 #14 of 22 Archive leaderboard report
Dichotomous Image Segmentation DIS-TE4 STDC max F-Measure 0.731 #14 of 22 Archive leaderboard report
Dichotomous Image Segmentation DIS-TE4 STDC weighted F-measure 0.652 #14 of 22 Archive leaderboard report
Dichotomous Image Segmentation DIS-VD STDC E-measure 0.817 #15 of 24 Archive leaderboard report
Dichotomous Image Segmentation DIS-VD STDC HCE 1598 #15 of 24 Archive leaderboard report
Dichotomous Image Segmentation DIS-VD STDC MAE 0.103 #15 of 24 Archive leaderboard report
Dichotomous Image Segmentation DIS-VD STDC S-Measure 0.740 #15 of 24 Archive leaderboard report
Dichotomous Image Segmentation DIS-VD STDC max F-Measure 0.696 #15 of 24 Archive leaderboard report
Dichotomous Image Segmentation DIS-VD STDC weighted F-measure 0.613 #15 of 24 Archive leaderboard report
Real-Time Semantic Segmentation Cityscapes test STDC2-75 Frame (fps) 97.0(1080Ti) #8 of 39 Archive leaderboard report
Real-Time Semantic Segmentation Cityscapes test STDC2-75 mIoU 76.8% #8 of 39 Archive leaderboard report
Real-Time Semantic Segmentation Cityscapes test STDC1-75 Frame (fps) 126.7 #13 of 39 Archive leaderboard report
Real-Time Semantic Segmentation Cityscapes test STDC1-75 mIoU 75.3% #13 of 39 Archive leaderboard report
Real-Time Semantic Segmentation Cityscapes test STDC2-50 Frame (fps) 188.6 #22 of 39 Archive leaderboard report
Real-Time Semantic Segmentation Cityscapes test STDC2-50 mIoU 73.4% #22 of 39 Archive leaderboard report
Real-Time Semantic Segmentation Cityscapes test STDC1-50 Frame (fps) 250.4(1080Ti) #25 of 39 Archive leaderboard report
Real-Time Semantic Segmentation Cityscapes test STDC1-50 mIoU 71.9% #25 of 39 Archive leaderboard report
Real-Time Semantic Segmentation Cityscapes val STDC2-Seg75 Frame (fps) 97 #11 of 24 Archive leaderboard report
Real-Time Semantic Segmentation Cityscapes val STDC2-Seg75 mIoU 77% #11 of 24 Archive leaderboard report
Real-Time Semantic Segmentation Cityscapes val STDC1-Seg75 Frame (fps) 126.7 #19 of 24 Archive leaderboard report
Real-Time Semantic Segmentation Cityscapes val STDC1-Seg75 mIoU 74.5% #19 of 24 Archive leaderboard report
Semantic Segmentation BDD100K val STDC1 mIoU 52.1(45.8FPS) #17 of 24 Archive leaderboard report
Semantic Segmentation BDD100K val STDC2 mIoU 53.8(33.0FPS) #18 of 24 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

Introduced by this paper: STDC

1x1 ConvolutionBatch NormalizationConvolutionReLUSTDC

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