{"url":"/sota/semantic-segmentation-on-bdd100k-val","task":{"name":"Semantic Segmentation","url":"/task/semantic-segmentation","note":null},"dataset":{"name":"BDD100K val","url":"/dataset/bdd100k"},"category":"Computer Vision","categories":["Computer Code","Computer Vision","Medical","Robots"],"category_note":null,"description":null,"description_from":null,"source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","rank":"the archive's row order at snapshot; not re-ranked","rows_end_at":"2025-07-28","rows_withheld_as_spam":0,"metric_values":"the archive's strings, untouched"},"metrics":["mIoU"],"metric_direction":{"note":"inferred from the metric name only (the archive records no direction); null = not inferred, chart draws points only","by_metric":{"mIoU":null}},"counts":{"rows":24,"rows_with_code":24,"rows_with_paper_page":24,"rows_dated":24,"rows_using_additional_data":1},"rows":[{"rank_in_archive_order":1,"model":"VLTSeg","metrics":{"mIoU":"72.5"},"uses_additional_data":false,"paper_date":"2023-12-04","paper":"/paper/vltseg-simple-transfer-of-clip-based-vision","paper_url":"https://arxiv.org/abs/2312.02021v4","paper_title":"Strong but simple: A Baseline for Domain Generalized Dense Perception by CLIP-based Transfer Learning","code":"https://github.com/VLTSeg/VLTSeg","n_code_links":1,"syntology":null},{"rank_in_archive_order":2,"model":"SERNet-Former_v2","metrics":{"mIoU":"67.42"},"uses_additional_data":true,"paper_date":"2024-01-28","paper":"/paper/sernet-former-semantic-segmentation-by","paper_url":"https://arxiv.org/abs/2401.15741v7","paper_title":"SERNet-Former: Semantic Segmentation by Efficient Residual Network with Attention-Boosting Gates and Attention-Fusion Networks","code":"https://github.com/serdarch/sernet-former","n_code_links":2,"syntology":null},{"rank_in_archive_order":3,"model":"DSNet-Base","metrics":{"mIoU":"64.6"},"uses_additional_data":false,"paper_date":"2024-06-06","paper":"/paper/dsnet-a-novel-way-to-use-atrous-convolutions","paper_url":"https://arxiv.org/abs/2406.03702v1","paper_title":"DSNet: A Novel Way to Use Atrous Convolutions in Semantic Segmentation","code":"https://github.com/takaniwa/dsnet","n_code_links":1,"syntology":null},{"rank_in_archive_order":4,"model":"Deeplabv3+","metrics":{"mIoU":"63.6"},"uses_additional_data":false,"paper_date":"2018-02-07","paper":"/paper/encoder-decoder-with-atrous-separable","paper_url":"http://arxiv.org/abs/1802.02611v3","paper_title":"Encoder-Decoder with Atrous Separable Convolution for Semantic Image Segmentation","code":"https://github.com/tensorflow/models/tree/master/research/deeplab","n_code_links":78,"syntology":{"n_ran":44,"n_unverified":28,"n_samples":72,"n_pointer_only_licence":40}},{"rank_in_archive_order":5,"model":"DANet","metrics":{"mIoU":"62.8"},"uses_additional_data":false,"paper_date":"2018-09-09","paper":"/paper/dual-attention-network-for-scene-segmentation","paper_url":"http://arxiv.org/abs/1809.02983v4","paper_title":"Dual Attention Network for Scene Segmentation","code":"https://github.com/xmu-xiaoma666/External-Attention-pytorch","n_code_links":12,"syntology":{"n_ran":0,"n_unverified":7,"n_samples":7,"n_pointer_only_licence":5}},{"rank_in_archive_order":6,"model":"PSPNet","metrics":{"mIoU":"62.3"},"uses_additional_data":false,"paper_date":"2016-12-04","paper":"/paper/pyramid-scene-parsing-network","paper_url":"http://arxiv.org/abs/1612.01105v2","paper_title":"Pyramid Scene Parsing Network","code":"https://github.com/tensorflow/models","n_code_links":67,"syntology":{"n_ran":7,"n_unverified":22,"n_samples":29,"n_pointer_only_licence":5}},{"rank_in_archive_order":7,"model":"EMANet","metrics":{"mIoU":"61.4"},"uses_additional_data":false,"paper_date":"2019-07-31","paper":"/paper/expectation-maximization-attention-networks","paper_url":"https://arxiv.org/abs/1907.13426v2","paper_title":"Expectation-Maximization Attention Networks for Semantic Segmentation","code":"https://github.com/open-mmlab/mmsegmentation","n_code_links":5,"syntology":null},{"rank_in_archive_order":8,"model":"OCRNet","metrics":{"mIoU":"60.1"},"uses_additional_data":false,"paper_date":"2019-09-24","paper":"/paper/object-contextual-representations-for","paper_url":"https://arxiv.org/abs/1909.11065v6","paper_title":"Segmentation Transformer: Object-Contextual Representations for Semantic Segmentation","code":"https://github.com/open-mmlab/mmsegmentation","n_code_links":11,"syntology":{"n_ran":5,"n_unverified":4,"n_samples":9,"n_pointer_only_licence":1}},{"rank_in_archive_order":9,"model":"NiseNet","metrics":{"mIoU":"53.52"},"uses_additional_data":false,"paper_date":"2019-09-24","paper":"/paper/what-s-there-in-the-dark","paper_url":"https://ieeexplore.ieee.org/document/8803299","paper_title":"What's There in the Dark","code":"https://github.com/sauradip/night_image_semantic_segmentation","n_code_links":1,"syntology":null},{"rank_in_archive_order":10,"model":"MRFP+(Ours) Resnet50","metrics":{"mIoU":"39.55"},"uses_additional_data":false,"paper_date":"2023-11-30","paper":"/paper/mrfp-learning-generalizable-semantic","paper_url":"https://arxiv.org/abs/2311.18331v2","paper_title":"MRFP: Learning Generalizable Semantic Segmentation from Sim-2-Real with Multi-Resolution Feature Perturbation","code":"https://github.com/airl-iisc/MRFP","n_code_links":1,"syntology":null},{"rank_in_archive_order":11,"model":"Resnet50","metrics":{"mIoU":"31.44"},"uses_additional_data":false,"paper_date":"2023-11-30","paper":"/paper/mrfp-learning-generalizable-semantic","paper_url":"https://arxiv.org/abs/2311.18331v2","paper_title":"MRFP: Learning Generalizable Semantic Segmentation from Sim-2-Real with Multi-Resolution Feature Perturbation","code":"https://github.com/airl-iisc/MRFP","n_code_links":1,"syntology":null},{"rank_in_archive_order":12,"model":"BiSeNet-V1(ResNet-18)","metrics":{"mIoU":"53.8(45.1fps)"},"uses_additional_data":false,"paper_date":"2018-08-02","paper":"/paper/bisenet-bilateral-segmentation-network-for","paper_url":"http://arxiv.org/abs/1808.00897v1","paper_title":"BiSeNet: Bilateral Segmentation Network for Real-time Semantic Segmentation","code":"https://github.com/PaddlePaddle/PaddleSeg","n_code_links":21,"syntology":{"n_ran":7,"n_unverified":11,"n_samples":18,"n_pointer_only_licence":1}},{"rank_in_archive_order":13,"model":"ICNet","metrics":{"mIoU":"52.4(39.5fps)"},"uses_additional_data":false,"paper_date":"2017-04-27","paper":"/paper/icnet-for-real-time-semantic-segmentation-on","paper_url":"http://arxiv.org/abs/1704.08545v2","paper_title":"ICNet for Real-Time Semantic Segmentation on High-Resolution Images","code":"https://github.com/osmr/imgclsmob","n_code_links":18,"syntology":null},{"rank_in_archive_order":14,"model":"Bi-Align","metrics":{"mIoU":"53.4(42.1fps)"},"uses_additional_data":false,"paper_date":"2021-05-25","paper":"/paper/fast-and-accurate-scene-parsing-via-bi","paper_url":"https://arxiv.org/abs/2105.11651v1","paper_title":"Fast and Accurate Scene Parsing via Bi-direction Alignment Networks","code":"https://github.com/jojacola/BiAlignNet","n_code_links":1,"syntology":null},{"rank_in_archive_order":15,"model":"DF1-Seg","metrics":{"mIoU":"42.5(82.3fps)"},"uses_additional_data":false,"paper_date":"2019-03-09","paper":"/paper/partial-order-pruning-for-best-speedaccuracy","paper_url":"http://arxiv.org/abs/1903.03777v2","paper_title":"Partial Order Pruning: for Best Speed/Accuracy Trade-off in Neural Architecture Search","code":"https://github.com/lixincn2015/Partial-Order-Pruning","n_code_links":2,"syntology":null},{"rank_in_archive_order":16,"model":"DF2-Seg","metrics":{"mIoU":"47.8(53.4fps)"},"uses_additional_data":false,"paper_date":"2019-03-09","paper":"/paper/partial-order-pruning-for-best-speedaccuracy","paper_url":"http://arxiv.org/abs/1903.03777v2","paper_title":"Partial Order Pruning: for Best Speed/Accuracy Trade-off in Neural Architecture Search","code":"https://github.com/lixincn2015/Partial-Order-Pruning","n_code_links":2,"syntology":null},{"rank_in_archive_order":17,"model":"STDC1","metrics":{"mIoU":"52.1(45.8FPS)"},"uses_additional_data":false,"paper_date":"2021-04-27","paper":"/paper/rethinking-bisenet-for-real-time-semantic","paper_url":"https://arxiv.org/abs/2104.13188v1","paper_title":"Rethinking BiSeNet For Real-time Semantic Segmentation","code":"https://github.com/PaddlePaddle/PaddleSeg","n_code_links":6,"syntology":{"n_ran":9,"n_unverified":7,"n_samples":16,"n_pointer_only_licence":0}},{"rank_in_archive_order":18,"model":"STDC2","metrics":{"mIoU":"53.8(33.0FPS)"},"uses_additional_data":false,"paper_date":"2021-04-27","paper":"/paper/rethinking-bisenet-for-real-time-semantic","paper_url":"https://arxiv.org/abs/2104.13188v1","paper_title":"Rethinking BiSeNet For Real-time Semantic Segmentation","code":"https://github.com/PaddlePaddle/PaddleSeg","n_code_links":6,"syntology":{"n_ran":9,"n_unverified":7,"n_samples":16,"n_pointer_only_licence":0}},{"rank_in_archive_order":19,"model":"SFNet(DF2)","metrics":{"mIoU":"60.2(208FPS 4090)"},"uses_additional_data":false,"paper_date":"2020-02-24","paper":"/paper/semantic-flow-for-fast-and-accurate-scene","paper_url":"https://arxiv.org/abs/2002.10120v3","paper_title":"Semantic Flow for Fast and Accurate Scene Parsing","code":"https://github.com/PaddlePaddle/PaddleSeg","n_code_links":6,"syntology":{"n_ran":1,"n_unverified":7,"n_samples":8,"n_pointer_only_licence":1}},{"rank_in_archive_order":20,"model":"SFNet(DF1)","metrics":{"mIoU":"55.4(70.3fps)"},"uses_additional_data":false,"paper_date":"2020-02-24","paper":"/paper/semantic-flow-for-fast-and-accurate-scene","paper_url":"https://arxiv.org/abs/2002.10120v3","paper_title":"Semantic Flow for Fast and Accurate Scene Parsing","code":"https://github.com/PaddlePaddle/PaddleSeg","n_code_links":6,"syntology":{"n_ran":1,"n_unverified":7,"n_samples":8,"n_pointer_only_licence":1}},{"rank_in_archive_order":21,"model":"SFNet(ResNet-18)","metrics":{"mIoU":"60.6(132.5FPS 4090)"},"uses_additional_data":false,"paper_date":"2020-02-24","paper":"/paper/semantic-flow-for-fast-and-accurate-scene","paper_url":"https://arxiv.org/abs/2002.10120v3","paper_title":"Semantic Flow for Fast and Accurate Scene Parsing","code":"https://github.com/PaddlePaddle/PaddleSeg","n_code_links":6,"syntology":{"n_ran":1,"n_unverified":7,"n_samples":8,"n_pointer_only_licence":1}},{"rank_in_archive_order":22,"model":"SFNet-Lite(ResNet-18)","metrics":{"mIoU":"60.6(161.3FPS 4090)"},"uses_additional_data":false,"paper_date":"2022-07-10","paper":"/paper/sfnet-faster-accurate-and-domain-agnostic","paper_url":"https://arxiv.org/abs/2207.04415v1","paper_title":"SFNet: Faster, Accurate, and Domain Agnostic Semantic Segmentation via Semantic Flow","code":"https://github.com/lxtGH/SFSegNets","n_code_links":1,"syntology":null},{"rank_in_archive_order":23,"model":"SFNet-Lite(STDC2)","metrics":{"mIoU":"60.6(194.5FPS 4090)"},"uses_additional_data":false,"paper_date":"2022-07-10","paper":"/paper/sfnet-faster-accurate-and-domain-agnostic","paper_url":"https://arxiv.org/abs/2207.04415v1","paper_title":"SFNet: Faster, Accurate, and Domain Agnostic Semantic Segmentation via Semantic Flow","code":"https://github.com/lxtGH/SFSegNets","n_code_links":1,"syntology":null},{"rank_in_archive_order":24,"model":"DSNet-head64","metrics":{"mIoU":"62.6(172.2FPS 4090)"},"uses_additional_data":false,"paper_date":"2024-06-06","paper":"/paper/dsnet-a-novel-way-to-use-atrous-convolutions","paper_url":"https://arxiv.org/abs/2406.03702v1","paper_title":"DSNet: A Novel Way to Use Atrous Convolutions in Semantic Segmentation","code":"https://github.com/takaniwa/dsnet","n_code_links":1,"syntology":null}],"since_archive":{"claim":"Results that newer papers report for their own method, placed here by Syntology. A model pointed at the cell in the paper's own table; the number was read from that cell and checked against this leaderboard's metric, dataset, split and scale; an independent check that saw this leaderboard's other rows and every other leaderboard on the same dataset accepted it. Not reviewed by the paper's authors or by the archive's editors, and not ranked against the archive rows.","extraction_file_present":true,"measurement":{"test_papers":883,"papers_with_output":881,"judged_true":108,"judged":110,"wilson95_lower":0.9361,"measured_on":"2026-09-24","frozen_commit":"0e3de0df94"},"measurement_note":"blind adjudication of accepted entries on a held-out split of archive papers, rules frozen before the test","coverage":{"sentence":"Syntology has checked 6,885 of the 9,623 papers on this site that are newer than the archive; results from the others appear after they are checked.","complete":false,"papers_newer_than_archive":9623,"papers_checked":6885,"papers_extracted_not_yet_verified":0,"boards_without_verdict":2,"papers_not_yet_extracted":2737},"order":"newest first by month (arXiv date, else the arXiv-id month), then arXiv id descending","columns":[],"entries":[]},"syntology":{"read_at":"2026-09-25T09:33:49+00:00","claim":"Per row: N of M harvested code samples from that row's paper executed on a synthesized fixture; the other M-N are unverified. Not a reproduction of the row's number; not a correctness claim. n_pointer_only_licence counts samples the site points at rather than redistributes (a licence axis, independent of ran/unverified).","rows_with_graph_line":10,"rows_with_any_sample_ran":9,"distinct_papers_with_graph_line":7,"distinct_papers_with_any_sample_ran":6,"samples_over_distinct_papers":{"n_ran":73,"n_unverified":86,"n_samples":159,"n_pointer_only_licence":53,"note":"each paper (arXiv id) counted once, however many rows it is behind; this is the page-level figure"},"samples_row_weighted":{"n_ran":84,"n_unverified":107,"n_samples":191,"n_pointer_only_licence":55,"note":"row-weighted: a paper behind several rows is counted once per row; inflated relative to samples_over_distinct_papers by design, kept for readers summing the per-row syntology blocks"}}}