Papers › Fast-SCNN: Fast Semantic Segmentation Network

Fast-SCNN: Fast Semantic Segmentation Network

12 Feb 2019arXiv:1902.04502archive 2025-07-28

Rudra P. K. Poudel, Stephan Liwicki, Roberto Cipolla

The encoder-decoder framework is state-of-the-art for offline semantic image segmentation. Since the rise in autonomous systems, real-time computation is increasingly desirable. In this paper, we introduce fast segmentation convolutional neural network (Fast-SCNN), an above real-time semantic segmentation model on high resolution image data (1024x2048px) suited to efficient computation on embedded devices with low memory. Building on existing two-branch methods for fast segmentation, we introduce our `learning to downsample' module which computes low-level features for multiple resolution branches simultaneously. Our network combines spatial detail at high resolution with deep features extracted at lower resolution, yielding an accuracy of 68.0% mean intersection over union at 123.5 frames per second on Cityscapes. We also show that large scale pre-training is unnecessary. We thoroughly validate our metric in experiments with ImageNet pre-training and the coarse labeled data of Cityscapes. Finally, we show even faster computation with competitive results on subsampled inputs, without any network modifications.

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 32 code samples harvested from 3 repositories linked to this paper: 21 ran (of which 0 constructed an object rather than computing a result; 21 with no instrument failure: 0 honoured, 0 violated, 21 with no contract checked; 0 where Syntology's instrument failed) · 11 unverified. “Where Syntology's instrument failed” counts failures of Syntology's instrument, not of the code; here there are none.

By repository: community (archive-listed): 32 samples from 3 repositories, 21 ran (21 with no instrument failure; 0 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 6 of its repositories that Syntology harvested for other papers:

Levels written on 2026-09-24 (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.

DeepVoltaire/Fast-SCNN mentioned on GitHubpytorchMITSyntology: not harvested report
Eladamar/fast_scnn mentioned on GitHubpytorchSyntology: not harvested report
JonasFrey96/ASL mentioned on GitHubpytorchSyntology: not harvested report
Mind23-2/MindCode-44 mentioned on GitHubmindsporeSyntology: not harvested report
SkyWa7ch3r/ImageSegmentation mentioned on GitHubtfSyntology: not harvested report
SkyWa7ch3r/SceneSegmentation mentioned on GitHubtfSyntology: no sample linked to this paper (harvested for another paper). Syntology's graph links these samples to Encoder-Decoder with Atrous Separable Convolution for Semantic Image Segmentation (arXiv:1802.02611; that paper's own run record: 44 ran (of which 17 constructed an object rather than computing a result; 28 with no instrument failure: 2 honoured, 0 violated, 26 with no contract checked; 16 where Syntology's instrument failed) · 28 unverified); nothing checks that they implement this paper's method. report
Tramac/Fast-SCNN-pytorch mentioned on GitHubpytorchApache-2.0Syntology: 5 samples, 4 ran (0 instrument-failed) · 1 unverified report
VilledeMontreal/urban-segmentation mentioned on GitHubpytorchMITSyntology: not harvested report
ayushmankumar7/Fast-Segementation mentioned on GitHubtfGPL-3.0Syntology: not harvested report
dbaofd/solar-panels-detection mentioned on GitHubtfMITSyntology: 10 samples, 9 ran (0 instrument-failed) · 1 unverified report
kshitizrimal/Fast-SCNN mentioned on GitHubtfSyntology: not harvested report
minus31/Portrait-segmentation mentioned on GitHubtfMITSyntology: 17 samples, 8 ran (0 instrument-failed) · 9 unverified report
osmr/imgclsmob mentioned on GitHubmxnetMITSyntology: no sample linked to this paper (harvested for another paper). Syntology's graph links these samples to ResNeSt: Split-Attention Networks (arXiv:2004.08955; that paper's own run record: 28 ran (of which 0 constructed an object rather than computing a result; 21 with no instrument failure: 0 honoured, 0 violated, 21 with no contract checked; 7 where Syntology's instrument failed) · 20 unverified); nothing checks that they implement this paper's method. report
rachthree/fast_scnn mentioned on GitHubtfSyntology: not harvested report
rudolfsteiner/fast_scnn mentioned on GitHubtfSyntology: not harvested report
zh320/realtime-semantic-segmentation-pytorch mentioned on GitHubpytorchSyntology: no sample linked to this paper (harvested for another paper). Syntology's graph links these samples to CGNet: A Light-weight Context Guided Network for Semantic Segmentation (arXiv:1811.08201; that paper's own run record: 2 ran (of which 0 constructed an object rather than computing a result; 1 with no instrument failure: 0 honoured, 1 violated, 0 with no contract checked; 1 where Syntology's instrument failed) · 0 unverified), ESPNetv2: A Light-weight, Power Efficient, and General Purpose Convolutional Neural Network (arXiv:1811.11431; that paper's own run record: 4 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; 2 where Syntology's instrument failed) · 0 unverified), Lite-HRNet: A Lightweight High-Resolution Network (arXiv:2104.06403; that paper's own run record: 13 ran (of which 2 constructed an object rather than computing a result; 11 with no instrument failure: 0 honoured, 0 violated, 11 with no contract checked; 2 where Syntology's instrument failed) · 16 unverified) and 1 more (every one in this page's JSON twin); nothing checks that they implement this paper's method. report
2023-MindSpore-1/ms-code-213 mindsporeSyntology: not harvested report
MindSpore-paper-code-3/code10 mindsporeSyntology: not harvested report
MindSpore-paper-code-3/code3 mindsporeSyntology: not harvested report
MindSpore-paper-code-3/code7 mindsporeSyntology: not harvested report
MindSpore-paper-code-3/code8 mindsporeSyntology: 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
mindspore-ai/models mindsporeSyntology: no sample linked to this paper (harvested for another paper). Syntology's graph links these samples to AdaBin: Improving Binary Neural Networks with Adaptive Binary Sets (arXiv:2208.08084; 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
open-mmlab/mmsegmentation pytorchApache-2.0Syntology: no sample linked to this paper (harvested for another paper). Syntology's graph links these samples to Perceptual Artifacts Localization for Image Synthesis Tasks (arXiv:2310.05590; that paper's own run record: 6 ran (of which 0 constructed an object rather than computing a result; 2 with no instrument failure: 1 honoured, 1 violated, 0 with no contract checked; 4 where Syntology's instrument failed) · 2 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

32 samples harvested: 21 ran (of which 0 constructed an object rather than computing a result; 21 with no instrument failure: 0 honoured, 0 violated, 21 with no contract checked; 0 where Syntology's instrument failed) · 11 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.

21ran
11unverified

Licence: 0 of the 32 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 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.

Tramac/Fast-SCNN-pytorch community (archive-listed) 5 samples, 4 ran (0 instrument-failed) · 1 unverified
batch_intersection_union in 10 papers utils/metric.py ran fingerprinted Apache-2.0 (permissive) · b0028c3b51c18eda · report fixture: signature predict:ndarray/2/int, target:ndarray/2/int, nclass:int/0/int · returned tuple · level recorded 2026-09-24
batch_pix_accuracy in 12 papers utils/metric.py ran Apache-2.0 (permissive) · d98304f5d56cebc0 · report fixture: signature predict:ndarray/2/float, target:ndarray/2/float · returned tuple · level recorded 2026-09-24
get_color_pallete utils/visualize.py ran Apache-2.0 (permissive) · 1b78db99bae50239 · report fixture: signature npimg:ndarray/2/int, dataset:str/0/str · returned Image · level recorded 2026-09-24
pixelAccuracy utils/metric.py ran fingerprinted Apache-2.0 (permissive) · 5a4af1d12a8f04f0 · report fixture: signature imPred:ndarray/2/int, imLab:ndarray/2/int · returned tuple · level recorded 2026-09-24
get_fast_scnn models/fast_scnn.py unverified Apache-2.0 (permissive) · acb7752b94575ac5 · report
dbaofd/solar-panels-detection community (archive-listed) 10 samples, 9 ran (0 instrument-failed) · 1 unverified
bottleneck_block in 6 papers model_list/fast_scnn_0.py ran MIT (permissive) · 5694a1e205acecda · report fixture: returned EagerTensor · level recorded 2026-09-24
conv_block in 95 papers model_list/fast_scnn_0.py ran MIT (permissive) · 721b77a7a5de2adf · report fixture: signature inputs:KerasTensor/0/float, conv_type:str/0/str, kernel:int/0/int, kernel_size:int/0/int, strides:int/0/int, padding:str/0/str, relu:bool/0/bool · returned KerasTensor · level recorded 2026-09-24
conv_block_3 model_list/segnet_2.py ran MIT (permissive) · 69ebb6a685a47229 · report fixture: signature input_tensor:KerasTensor/0/float, kernel_size:int/0/int, filters:list/1/int, stage:int/0/int, block:str/0/str · returned KerasTensor · level recorded 2026-09-24
crop_image in 17 papers data_processing/data_processing_tool_1.py ran MIT (permissive) · fa7eb507337ce943 · report fixture: signature img:ndarray/3/float · returned Image · level recorded 2026-09-24
crop_image in 17 papers data_processing/data_processing_tool_2.py ran MIT (permissive) · 01bf9e1fd4ec484d · report fixture: signature img:ndarray/3/int · returned list · level recorded 2026-09-24
identity_block in 32 papers model_list/segnet_3.py ran MIT (permissive) · f73ed37393ddb228 · report fixture: signature input_tensor:KerasTensor/0/float, kernel_size:int/0/int, filters:list/1/int, stage:int/0/int, block:str/0/str, next_layer_filter:int/0/int · returned KerasTensor · level recorded 2026-09-24
identity_block_3 model_list/segnet_2.py ran MIT (permissive) · 45aae890a696aa1d · report fixture: signature input_tensor:KerasTensor/4/float, kernel_size:int/0/int, filters:list/1/int, stage:int/0/int, block:str/0/str, next_layer_filter:int/0/int · returned KerasTensor · level recorded 2026-09-24
pyramid_pooling_block model_list/fast_scnn_0.py ran MIT (permissive) · f61ffd42fa393081 · report fixture: signature input_tensor:EagerTensor/4/float, bin_sizes:list/1/int · returned EagerTensor · level recorded 2026-09-24
pyramid_pooling_block model_list/fast_scnn_1.py ran MIT (permissive) · a2fc73a1ddb8d26a · report fixture: signature input_tensor:SymbolicTensor/4/float, bin_sizes:list/1/int · returned EagerTensor · level recorded 2026-09-24
conv_block in 95 papers model_list/segnet_3.py unverified MIT (permissive) · 740cb4ae8f653627 · report
minus31/Portrait-segmentation community (archive-listed) 17 samples, 8 ran (0 instrument-failed) · 9 unverified
bilinear_interpolation fast_scnn.py ran MIT (permissive) · fe1f1ddb18f26347 · report fixture: signature x:EagerTensor/4/float, size:list/1/int · returned EagerTensor · level recorded 2026-09-24
composite4 indexnet/Composition_code.py ran MIT (permissive) · e41c7308aeed9225 · report fixture: signature fg:ndarray/3/int, bg:ndarray/3/int, a:ndarray/2/int, w:int/0/int, h:int/0/int · returned ndarray · level recorded 2026-09-24
dilated_conv_3x3_bn indexnet/hlaspp.py ran MIT (permissive) · b7f80e16cc196e71 · report fixture: signature inp:int/0/int, oup:int/0/int, padding:int/0/int, dilation:int/0/int, BatchNorm2d:callable/0/? · returned Sequential · level recorded 2026-09-24
hflip in 29 papers augmentation.py ran fingerprinted MIT (permissive) · 6c6cdac56400f7eb · report fixture: signature img:ndarray/2/float, mask:ndarray/2/float · returned tuple · level recorded 2026-09-24
identity_img augmentation.py ran MIT (permissive) · b88010e1312d8957 · report fixture: signature img:Tensor/4/float, mask:Tensor/3/int · returned tuple · level recorded 2026-09-24
load_img_mask_pair get_matte_mask_from_binary_mask.py ran MIT (permissive) · ffc64df0f93c1c55 · report fixture: signature img_path:str/0/str · returned tuple · level recorded 2026-09-24
make_trimap get_matte_mask_from_binary_mask.py ran MIT (permissive) · ebff3c686211b2ae · report fixture: signature mask:ndarray/2/int, size:tuple/1/int · returned ndarray · level recorded 2026-09-24
random_activate augmentation.py ran MIT (permissive) · 424755265e383fb2 · report fixture: signature aug_func:callable/0/? · returned function · level recorded 2026-09-24
depth_sep_dilated_conv_3x3_bn indexnet/hlaspp.py unverified MIT (permissive) · 145fc345406f0e26 · report
get_edge in 5 papers data_generator.py unverified MIT (permissive) · c35601c93ff05d52 · report
linearBottleneck fast_scnn.py unverified MIT (permissive) · 8686920e76a2f219 · report
network_big network.py unverified MIT (permissive) · 1924371588f8b45b · report
network_small network.py unverified MIT (permissive) · 2f1f371d25d86691 · report
read_image in 73 papers get_matte_mask_from_binary_mask.py unverified MIT (permissive) · 5eba125ca4ddd12a · report
residual_block in 18 papers network.py unverified MIT (permissive) · 40bf1a25c403d94a · report
rotate90_img data_generator.py unverified MIT (permissive) · 71303057584fee4a · report
swish_activation fast_scnn.py unverified MIT (permissive) · 29e46b6d9898edb6 · report

Tasks

DecoderImage SegmentationReal-Time Semantic SegmentationSegmentationSemantic SegmentationThermal Image Segmentation

Results from the paper archive 2025-07-28

TaskDatasetModelMetricValueRank at snapshotLeaderboardReport
Semantic Segmentation Cityscapes test Fast-SCNN Mean IoU (class) 68% #84 of 105 Archive leaderboard report
Semantic Segmentation Cityscapes val Fast-SCNN + Coarse + ImageNet mIoU 69.19 #88 of 99 Archive leaderboard report
Semantic Segmentation DADA-seg Fast-SCNN mIoU 26.32 #13 of 28 Archive leaderboard report
Semantic Segmentation DensePASS Fast-SCNN mIoU 24.6% #34 of 36 Archive leaderboard report
Semantic Segmentation EventScape Fast-SCNN mIoU 44.27 #9 of 12 Archive leaderboard report
Semantic Segmentation SynPASS Fast-SCNN mIoU 21.30% #6 of 6 Archive leaderboard report
Thermal Image Segmentation PST900 Fast-SCNN mIoU 47.2 #22 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.

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