{"url":"/sota/domain-adaptation-on-synthia-to-cityscapes","task":{"name":"Domain Adaptation","url":"/task/domain-adaptation","note":null},"dataset":{"name":"SYNTHIA-to-Cityscapes","url":"/dataset/synthia"},"category":"Computer Vision","categories":["Computer Vision","Methodology"],"category_note":null,"description":"**Domain Adaptation** is the task of adapting models across domains. This is motivated by the challenge where the test and training datasets fall from different data distributions due to some factor. Domain adaptation aims to build machine learning models that can be generalized into a target domain and dealing with the discrepancy across domain distributions. \r\n\r\nFurther readings:\r\n\r\n- [A Brief Review of Domain Adaptation](https://paperswithcode.com/paper/a-brief-review-of-domain-adaptation)\r\n\r\n<span style=\"color:grey; opacity: 0.6\">( Image credit: [Unsupervised Image-to-Image Translation Networks](https://arxiv.org/pdf/1703.00848v6.pdf) )</span>","description_from":"task","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","Extra Manual Annotation"],"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,"Extra Manual Annotation":null}},"counts":{"rows":33,"rows_with_code":26,"rows_with_paper_page":33,"rows_dated":33,"rows_using_additional_data":2},"rows":[{"rank_in_archive_order":1,"model":"HALO","metrics":{"Extra Manual Annotation":"Yes","mIoU":"78.1"},"uses_additional_data":false,"paper_date":"2023-06-19","paper":"/paper/hyperbolic-active-learning-for-semantic","paper_url":"https://arxiv.org/abs/2306.11180v5","paper_title":"Hyperbolic Active Learning for Semantic Segmentation under Domain Shift","code":"https://github.com/paolomandica/HALO","n_code_links":1,"syntology":{"n_ran":3,"n_unverified":5,"n_samples":8,"n_pointer_only_licence":0}},{"rank_in_archive_order":2,"model":"ILM-ASSL","metrics":{"Extra Manual Annotation":"Yes","mIoU":"76.6"},"uses_additional_data":false,"paper_date":"2023-01-31","paper":"/paper/iterative-loop-learning-combining-self","paper_url":"https://arxiv.org/abs/2301.13361v4","paper_title":"Iterative Loop Method Combining Active and Semi-Supervised Learning for Domain Adaptive Semantic Segmentation","code":"https://github.com/licongguan/ilm-assl","n_code_links":1,"syntology":null},{"rank_in_archive_order":3,"model":"DCF","metrics":{"mIoU":"69.3"},"uses_additional_data":false,"paper_date":"2023-11-21","paper":"/paper/transferring-to-real-world-layouts-a-depth","paper_url":"https://arxiv.org/abs/2311.12682v1","paper_title":"Transferring to Real-World Layouts: A Depth-aware Framework for Scene Adaptation","code":"https://github.com/chen742/PiPa","n_code_links":2,"syntology":null},{"rank_in_archive_order":4,"model":"HRDA+PiPa","metrics":{"mIoU":"68.2"},"uses_additional_data":false,"paper_date":"2022-11-14","paper":"/paper/pipa-pixel-and-patch-wise-self-supervised","paper_url":"https://arxiv.org/abs/2211.07609v1","paper_title":"PiPa: Pixel- and Patch-wise Self-supervised Learning for Domain Adaptative Semantic Segmentation","code":"https://github.com/chen742/PiPa","n_code_links":1,"syntology":null},{"rank_in_archive_order":5,"model":"MIC","metrics":{"mIoU":"67.3"},"uses_additional_data":false,"paper_date":"2022-12-02","paper":"/paper/mic-masked-image-consistency-for-context","paper_url":"https://arxiv.org/abs/2212.01322v2","paper_title":"MIC: Masked Image Consistency for Context-Enhanced Domain Adaptation","code":"https://github.com/lhoyer/mic","n_code_links":1,"syntology":null},{"rank_in_archive_order":6,"model":"FREDOM - Transformer","metrics":{"mIoU":"67"},"uses_additional_data":false,"paper_date":"2023-04-04","paper":"/paper/fredom-fairness-domain-adaptation-approach-to","paper_url":"https://arxiv.org/abs/2304.02135v1","paper_title":"FREDOM: Fairness Domain Adaptation Approach to Semantic Scene Understanding","code":"https://github.com/uark-cviu/fredom","n_code_links":1,"syntology":{"n_ran":4,"n_unverified":0,"n_samples":4,"n_pointer_only_licence":4}},{"rank_in_archive_order":7,"model":"HRDA","metrics":{"mIoU":"65.8"},"uses_additional_data":false,"paper_date":"2022-04-27","paper":"/paper/hrda-context-aware-high-resolution-domain","paper_url":"https://arxiv.org/abs/2204.13132v2","paper_title":"HRDA: Context-Aware High-Resolution Domain-Adaptive Semantic Segmentation","code":"https://github.com/lhoyer/hrda","n_code_links":1,"syntology":null},{"rank_in_archive_order":8,"model":"SePiCo","metrics":{"mIoU":"64.3"},"uses_additional_data":false,"paper_date":"2022-04-19","paper":"/paper/sepico-semantic-guided-pixel-contrast-for","paper_url":"https://arxiv.org/abs/2204.08808v2","paper_title":"SePiCo: Semantic-Guided Pixel Contrast for Domain Adaptive Semantic Segmentation","code":"https://github.com/bit-da/sepico","n_code_links":1,"syntology":{"n_ran":1,"n_unverified":0,"n_samples":1,"n_pointer_only_licence":1}},{"rank_in_archive_order":9,"model":"MIC + Guidance Training","metrics":{"mIoU":"63.8"},"uses_additional_data":false,"paper_date":"2024-03-22","paper":"/paper/improve-cross-domain-mixed-sampling-with","paper_url":"https://arxiv.org/abs/2403.14995v1","paper_title":"Improve Cross-domain Mixed Sampling with Guidance Training for Adaptive Segmentation","code":"https://github.com/wenlve-zhou/guidance-training","n_code_links":1,"syntology":null},{"rank_in_archive_order":10,"model":"DAFormer + ProCST","metrics":{"mIoU":"61.6"},"uses_additional_data":true,"paper_date":"2022-04-25","paper":"/paper/procst-boosting-semantic-segmentation-using","paper_url":"https://arxiv.org/abs/2204.11891v2","paper_title":"ProCST: Boosting Semantic Segmentation Using Progressive Cyclic Style-Transfer","code":"https://github.com/shahaf1313/procst","n_code_links":1,"syntology":null},{"rank_in_archive_order":11,"model":"DAFormer","metrics":{"mIoU":"60.9"},"uses_additional_data":false,"paper_date":"2021-11-29","paper":"/paper/daformer-improving-network-architectures-and","paper_url":"https://arxiv.org/abs/2111.14887v2","paper_title":"DAFormer: Improving Network Architectures and Training Strategies for Domain-Adaptive Semantic Segmentation","code":"https://github.com/lhoyer/DAFormer","n_code_links":3,"syntology":{"n_ran":6,"n_unverified":0,"n_samples":6,"n_pointer_only_licence":6}},{"rank_in_archive_order":12,"model":"GtA-SFDA (DeepLabv2-ResNet101)","metrics":{"mIoU":"60.1"},"uses_additional_data":false,"paper_date":"2021-08-25","paper":"/paper/generalize-then-adapt-source-free-domain","paper_url":"https://arxiv.org/abs/2108.11249v1","paper_title":"Generalize then Adapt: Source-Free Domain Adaptive Semantic Segmentation","code":"https://github.com/val-iisc/SFDA-Seg","n_code_links":1,"syntology":null},{"rank_in_archive_order":13,"model":"FREDOM - DeepLabV2","metrics":{"mIoU":"59.1"},"uses_additional_data":false,"paper_date":"2023-04-04","paper":"/paper/fredom-fairness-domain-adaptation-approach-to","paper_url":"https://arxiv.org/abs/2304.02135v1","paper_title":"FREDOM: Fairness Domain Adaptation Approach to Semantic Scene Understanding","code":"https://github.com/uark-cviu/fredom","n_code_links":1,"syntology":{"n_ran":4,"n_unverified":0,"n_samples":4,"n_pointer_only_licence":4}},{"rank_in_archive_order":14,"model":"SePiCo (DeepLabv2-ResNet-101)","metrics":{"mIoU":"58.1"},"uses_additional_data":false,"paper_date":"2022-04-19","paper":"/paper/sepico-semantic-guided-pixel-contrast-for","paper_url":"https://arxiv.org/abs/2204.08808v2","paper_title":"SePiCo: Semantic-Guided Pixel Contrast for Domain Adaptive Semantic Segmentation","code":"https://github.com/bit-da/sepico","n_code_links":1,"syntology":{"n_ran":1,"n_unverified":0,"n_samples":1,"n_pointer_only_licence":1}},{"rank_in_archive_order":15,"model":"ProDA+CRA","metrics":{"mIoU":"56.9"},"uses_additional_data":false,"paper_date":"2021-09-14","paper":"/paper/cross-region-domain-adaptation-for-class","paper_url":"https://arxiv.org/abs/2109.06422v2","paper_title":"Cross-Region Domain Adaptation for Class-level Alignment","code":null,"n_code_links":0,"syntology":null},{"rank_in_archive_order":16,"model":"CorDA (ResNet-101)","metrics":{"mIoU":"55.0"},"uses_additional_data":true,"paper_date":"2021-04-28","paper":"/paper/domain-adaptive-semantic-segmentation-with","paper_url":"https://arxiv.org/abs/2104.13613v2","paper_title":"Domain Adaptive Semantic Segmentation with Self-Supervised Depth Estimation","code":"https://github.com/qinenergy/corda","n_code_links":1,"syntology":{"n_ran":3,"n_unverified":0,"n_samples":3,"n_pointer_only_licence":3}},{"rank_in_archive_order":17,"model":"SAC (ResNet-101)","metrics":{"mIoU":"52.6"},"uses_additional_data":false,"paper_date":"2021-04-30","paper":"/paper/self-supervised-augmentation-consistency-for","paper_url":"https://arxiv.org/abs/2105.00097v1","paper_title":"Self-supervised Augmentation Consistency for Adapting Semantic Segmentation","code":"https://github.com/visinf/da-sac","n_code_links":1,"syntology":{"n_ran":2,"n_unverified":1,"n_samples":3,"n_pointer_only_licence":0}},{"rank_in_archive_order":18,"model":"STPL","metrics":{"mIoU":"51.8"},"uses_additional_data":false,"paper_date":"2023-03-25","paper":"/paper/spatio-temporal-pixel-level-contrastive","paper_url":"https://arxiv.org/abs/2303.14361v1","paper_title":"Spatio-Temporal Pixel-Level Contrastive Learning-based Source-Free Domain Adaptation for Video Semantic Segmentation","code":"https://github.com/shaoyuanlo/stpl","n_code_links":1,"syntology":null},{"rank_in_archive_order":19,"model":"RPT (ResNet-101)","metrics":{"mIoU":"51.2"},"uses_additional_data":false,"paper_date":"2020-06-11","paper":"/paper/transferring-and-regularizing-prediction-for-1","paper_url":"https://arxiv.org/abs/2006.06570v1","paper_title":"Transferring and Regularizing Prediction for Semantic Segmentation","code":null,"n_code_links":0,"syntology":null},{"rank_in_archive_order":20,"model":"IAST (ResNet-101)","metrics":{"mIoU":"49.8"},"uses_additional_data":false,"paper_date":"2020-08-27","paper":"/paper/instance-adaptive-self-training-for","paper_url":"https://arxiv.org/abs/2008.12197v1","paper_title":"Instance Adaptive Self-Training for Unsupervised Domain Adaptation","code":"https://github.com/bupt-ai-cz/IAST-ECCV2020","n_code_links":2,"syntology":null},{"rank_in_archive_order":21,"model":"SAC (VGG-16)","metrics":{"mIoU":"49.1"},"uses_additional_data":false,"paper_date":"2021-04-30","paper":"/paper/self-supervised-augmentation-consistency-for","paper_url":"https://arxiv.org/abs/2105.00097v1","paper_title":"Self-supervised Augmentation Consistency for Adapting Semantic Segmentation","code":"https://github.com/visinf/da-sac","n_code_links":1,"syntology":{"n_ran":2,"n_unverified":1,"n_samples":3,"n_pointer_only_licence":0}},{"rank_in_archive_order":22,"model":"PyCDA (ResNet-101)","metrics":{"mIoU":"46.7"},"uses_additional_data":false,"paper_date":"2019-08-26","paper":"/paper/constructing-self-motivated-pyramid","paper_url":"https://arxiv.org/abs/1908.09547v1","paper_title":"Constructing Self-motivated Pyramid Curriculums for Cross-Domain Semantic Segmentation: A Non-Adversarial Approach","code":"https://github.com/lianqing11/pycda","n_code_links":1,"syntology":null},{"rank_in_archive_order":23,"model":"BiMaL","metrics":{"mIoU":"46.2"},"uses_additional_data":false,"paper_date":"2021-08-06","paper":"/paper/bimal-bijective-maximum-likelihood-approach","paper_url":"https://arxiv.org/abs/2108.03267v1","paper_title":"BiMaL: Bijective Maximum Likelihood Approach to Domain Adaptation in Semantic Scene Segmentation","code":"https://github.com/uark-cviu/bimal","n_code_links":1,"syntology":null},{"rank_in_archive_order":24,"model":"FADA (ResNet-101)","metrics":{"mIoU":"45.2"},"uses_additional_data":false,"paper_date":"2020-07-17","paper":"/paper/classes-matter-a-fine-grained-adversarial","paper_url":"https://arxiv.org/abs/2007.09222v1","paper_title":"Classes Matter: A Fine-grained Adversarial Approach to Cross-domain Semantic Segmentation","code":"https://github.com/JDAI-CV/FADA","n_code_links":1,"syntology":null},{"rank_in_archive_order":25,"model":"PIT (ResNet-101)","metrics":{"mIoU":"44.0"},"uses_additional_data":false,"paper_date":"2020-06-01","paper":"/paper/cross-domain-semantic-segmentation-via-domain","paper_url":"http://openaccess.thecvf.com/content_CVPR_2020/html/Lv_Cross-Domain_Semantic_Segmentation_via_Domain-Invariant_Interactive_Relation_Transfer_CVPR_2020_paper.html","paper_title":"Cross-Domain Semantic Segmentation via Domain-Invariant Interactive Relation Transfer","code":null,"n_code_links":0,"syntology":null},{"rank_in_archive_order":26,"model":"SA-I2I (VGG-16)","metrics":{"mIoU":"41.5"},"uses_additional_data":false,"paper_date":"2020-09-02","paper":"/paper/semantically-adaptive-image-to-image","paper_url":"https://arxiv.org/abs/2009.01166v1","paper_title":"Semantically Adaptive Image-to-image Translation for Domain Adaptation of Semantic Segmentation","code":null,"n_code_links":0,"syntology":null},{"rank_in_archive_order":27,"model":"ADVENT (ResNet-101)","metrics":{"mIoU":"41.2"},"uses_additional_data":false,"paper_date":"2018-11-30","paper":"/paper/advent-adversarial-entropy-minimization-for","paper_url":"http://arxiv.org/abs/1811.12833v2","paper_title":"ADVENT: Adversarial Entropy Minimization for Domain Adaptation in Semantic Segmentation","code":"https://github.com/thuml/Transfer-Learning-Library","n_code_links":4,"syntology":{"n_ran":1,"n_unverified":2,"n_samples":3,"n_pointer_only_licence":1}},{"rank_in_archive_order":28,"model":"LDR (VGG-16)","metrics":{"mIoU":"41.1"},"uses_additional_data":false,"paper_date":"2020-03-10","paper":"/paper/label-driven-reconstruction-for-domain","paper_url":"https://arxiv.org/abs/2003.04614v3","paper_title":"Label-Driven Reconstruction for Domain Adaptation in Semantic Segmentation","code":null,"n_code_links":0,"syntology":null},{"rank_in_archive_order":29,"model":"CD-AM (VGG-16)","metrics":{"mIoU":"40.8"},"uses_additional_data":false,"paper_date":"2020-03-09","paper":"/paper/context-aware-domain-adaptation-in-semantic","paper_url":"https://arxiv.org/abs/2003.04010v1","paper_title":"Context-Aware Domain Adaptation in Semantic Segmentation","code":null,"n_code_links":0,"syntology":null},{"rank_in_archive_order":30,"model":"FDA (VGG-16)","metrics":{"mIoU":"40.5"},"uses_additional_data":false,"paper_date":"2020-04-11","paper":"/paper/fda-fourier-domain-adaptation-for-semantic","paper_url":"https://arxiv.org/abs/2004.05498v1","paper_title":"FDA: Fourier Domain Adaptation for Semantic Segmentation","code":"https://github.com/albumentations-team/albumentations","n_code_links":3,"syntology":{"n_ran":2,"n_unverified":1,"n_samples":3,"n_pointer_only_licence":3}},{"rank_in_archive_order":31,"model":"FADA (VGG-16)","metrics":{"mIoU":"39.5"},"uses_additional_data":false,"paper_date":"2020-07-17","paper":"/paper/classes-matter-a-fine-grained-adversarial","paper_url":"https://arxiv.org/abs/2007.09222v1","paper_title":"Classes Matter: A Fine-grained Adversarial Approach to Cross-domain Semantic Segmentation","code":"https://github.com/JDAI-CV/FADA","n_code_links":1,"syntology":null},{"rank_in_archive_order":32,"model":"PIT (VGG-16)","metrics":{"mIoU":"38.1"},"uses_additional_data":false,"paper_date":"2020-06-01","paper":"/paper/cross-domain-semantic-segmentation-via-domain","paper_url":"http://openaccess.thecvf.com/content_CVPR_2020/html/Lv_Cross-Domain_Semantic_Segmentation_via_Domain-Invariant_Interactive_Relation_Transfer_CVPR_2020_paper.html","paper_title":"Cross-Domain Semantic Segmentation via Domain-Invariant Interactive Relation Transfer","code":null,"n_code_links":0,"syntology":null},{"rank_in_archive_order":33,"model":"PyCDA (VGG-16)","metrics":{"mIoU":"35.9"},"uses_additional_data":false,"paper_date":"2019-08-26","paper":"/paper/constructing-self-motivated-pyramid","paper_url":"https://arxiv.org/abs/1908.09547v1","paper_title":"Constructing Self-motivated Pyramid Curriculums for Cross-Domain Semantic Segmentation: A Non-Adversarial Approach","code":"https://github.com/lianqing11/pycda","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 7,081 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":7081,"papers_extracted_not_yet_verified":217,"boards_without_verdict":27,"papers_not_yet_extracted":2325},"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":11,"rows_with_any_sample_ran":11,"distinct_papers_with_graph_line":8,"distinct_papers_with_any_sample_ran":8,"samples_over_distinct_papers":{"n_ran":22,"n_unverified":9,"n_samples":31,"n_pointer_only_licence":18,"note":"each paper (arXiv id) counted once, however many rows it is behind; this is the page-level figure"},"samples_row_weighted":{"n_ran":29,"n_unverified":10,"n_samples":39,"n_pointer_only_licence":23,"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"}}}