{"about":{"site":"https://codewithpapers.app","non_affiliation":"Code with Papers and Syntology are not affiliated with, endorsed by, or sponsored by Papers with Code, Meta, or the pwc-archive mirror.","licence":"CC BY-SA 4.0","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","attribution":"https://codewithpapers.app/attribution","modified":"archive material modified by Syntology; see the attribution page"},"url":"/paper/a-curriculum-domain-adaptation-approach-to","title":"A Curriculum Domain Adaptation Approach to the Semantic Segmentation of Urban Scenes","arxiv_id":"1812.09953","date":"2018-12-24","proceeding":null,"authors":["Yang Zhang","Philip David","Hassan Foroosh","Boqing Gong"],"abstract":"During the last half decade, convolutional neural networks (CNNs) have\ntriumphed over semantic segmentation, which is one of the core tasks in many\napplications such as autonomous driving and augmented reality. However, to\ntrain CNNs requires a considerable amount of data, which is difficult to\ncollect and laborious to annotate. Recent advances in computer graphics make it\npossible to train CNNs on photo-realistic synthetic imagery with\ncomputer-generated annotations. Despite this, the domain mismatch between the\nreal images and the synthetic data hinders the models' performance. Hence, we\npropose a curriculum-style learning approach to minimizing the domain gap in\nurban scene semantic segmentation. The curriculum domain adaptation solves easy\ntasks first to infer necessary properties about the target domain; in\nparticular, the first task is to learn global label distributions over images\nand local distributions over landmark superpixels. These are easy to estimate\nbecause images of urban scenes have strong idiosyncrasies (e.g., the size and\nspatial relations of buildings, streets, cars, etc.). We then train a\nsegmentation network, while regularizing its predictions in the target domain\nto follow those inferred properties. In experiments, our method outperforms the\nbaselines on two datasets and two backbone networks. We also report extensive\nablation studies about our approach.","url_abs":"http://arxiv.org/abs/1812.09953v3","url_pdf":"http://arxiv.org/pdf/1812.09953v3.pdf","source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","row_kind":"abstracts"},"code_links":[{"paper_slug":"a-curriculum-domain-adaptation-approach-to","repo_url":"https://github.com/YangZhang4065/AdaptationSeg","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"tf","reach":null},{"paper_slug":"a-curriculum-domain-adaptation-approach-to","repo_url":"https://github.com/mathilde-b/SRDA","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok"}}],"tasks":[{"task_slug":"autonomous-driving","task_name":"Autonomous Driving"},{"task_slug":"domain-adaptation","task_name":"Domain Adaptation"},{"task_slug":"image-to-image-translation","task_name":"Image-to-Image Translation"},{"task_slug":"segmentation","task_name":"Segmentation"},{"task_slug":"semantic-segmentation","task_name":"Semantic Segmentation"},{"task_slug":"superpixels","task_name":"Superpixels"},{"task_slug":"synthetic-to-real-translation","task_name":"Synthetic-to-Real Translation"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[{"leaderboard":"/sota/image-to-image-translation-on-synthia-to","task":"Image-to-Image Translation","dataset":"SYNTHIA-to-Cityscapes","model":"superpixel + color constancy","rank_in_archive_order":26,"of":28,"metrics":{"mIoU (13 classes)":"29.7"},"uses_additional_data":false},{"leaderboard":"/sota/synthetic-to-real-translation-on-gtav-to","task":"Synthetic-to-Real Translation","dataset":"GTAV-to-Cityscapes Labels","model":"superpixel + color constancy","rank_in_archive_order":71,"of":73,"metrics":{"mIoU":"31.4"},"uses_additional_data":false}],"syntology":{"syntology_url":"https://syntology.ai/paper/1812.09953","atlas_url":"https://app.syntology.ai/?focus=1812.09953","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"1812.09953"}},"developers":"https://syntology.ai/developers","read_at":"2026-09-25T09:33:49+00:00","read_at_is":"when the build read Syntology's graph, not when any sample ran","claim":"Per-sample execution status on synthesized fixtures; not a correctness claim about the paper. Samples come from repositories linked to the paper, official or community; repo_kind says which.","repos":[{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/mathilde-b/SRDA","reach":{"status":"ok"}},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/YangZhang4065/AdaptationSeg","reach":null}],"summary":{"unverified":1},"by_repo_kind":{"official":{"samples":1,"ran":0,"repositories":1}},"repo_kind_vocabulary":{"official":"The archive marks this repository official for the paper","named_in_paper":"The archive records that the paper mentions this repository; it is not marked official","listed":"In the archive's code links for this paper, not marked official and not recorded as mentioned in the paper","found_in_text":"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"},"n_pointer_only_for_licence":1,"samples":[{"code_sha256_prefix":"c707d2609c74b8a1","entry":"binarize_label","repo":"YangZhang4065/AdaptationSeg","repo_kind":"official","path":"train_val_FCN_DA.py","file_url":"https://github.com/YangZhang4065/AdaptationSeg/blob/HEAD/train_val_FCN_DA.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"c707d2609c74b8a1"}}]},"arxiv_metadata":null,"syntology_extracted_results":null}