{"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/auto-deeplab-hierarchical-neural-architecture","title":"Auto-DeepLab: Hierarchical Neural Architecture Search for Semantic Image Segmentation","arxiv_id":"1901.02985","date":"2019-01-10","proceeding":"CVPR 2019 6","authors":["Chenxi Liu","Liang-Chieh Chen","Florian Schroff","Hartwig Adam","Wei Hua","Alan Yuille","Li Fei-Fei"],"abstract":"Recently, Neural Architecture Search (NAS) has successfully identified neural\nnetwork architectures that exceed human designed ones on large-scale image\nclassification. In this paper, we study NAS for semantic image segmentation.\nExisting works often focus on searching the repeatable cell structure, while\nhand-designing the outer network structure that controls the spatial resolution\nchanges. This choice simplifies the search space, but becomes increasingly\nproblematic for dense image prediction which exhibits a lot more network level\narchitectural variations. Therefore, we propose to search the network level\nstructure in addition to the cell level structure, which forms a hierarchical\narchitecture search space. We present a network level search space that\nincludes many popular designs, and develop a formulation that allows efficient\ngradient-based architecture search (3 P100 GPU days on Cityscapes images). We\ndemonstrate the effectiveness of the proposed method on the challenging\nCityscapes, PASCAL VOC 2012, and ADE20K datasets. Auto-DeepLab, our\narchitecture searched specifically for semantic image segmentation, attains\nstate-of-the-art performance without any ImageNet pretraining.","url_abs":"http://arxiv.org/abs/1901.02985v2","url_pdf":"http://arxiv.org/pdf/1901.02985v2.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":"auto-deeplab-hierarchical-neural-architecture","repo_url":"https://github.com/tensorflow/models","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":0,"framework":"tf","reach":null},{"paper_slug":"auto-deeplab-hierarchical-neural-architecture","repo_url":"https://github.com/Dawars/auto_deeplab-pytorch","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":null},{"paper_slug":"auto-deeplab-hierarchical-neural-architecture","repo_url":"https://github.com/MenghaoGuo/AutoDeeplab","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok"}},{"paper_slug":"auto-deeplab-hierarchical-neural-architecture","repo_url":"https://github.com/Mind23-2/MindCode-26","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"mindspore","reach":{"status":"ok"}},{"paper_slug":"auto-deeplab-hierarchical-neural-architecture","repo_url":"https://github.com/NoamRosenberg/autodeeplab","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"gone","observed_at":"2026-09-18","how":"tree_404+repo_404"}},{"paper_slug":"auto-deeplab-hierarchical-neural-architecture","repo_url":"https://github.com/zhangximing666/Auto-DeepLab_MS","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"mindspore","reach":{"status":"ok","spdx":"MIT"}},{"paper_slug":"auto-deeplab-hierarchical-neural-architecture","repo_url":"https://github.com/2023-MindSpore-1/ms-code-16/tree/main/Auto-DeepLab","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":0,"framework":"mindspore","reach":null},{"paper_slug":"auto-deeplab-hierarchical-neural-architecture","repo_url":"https://github.com/2023-MindSpore-4/Code8/tree/main/Auto-DeepLab","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":0,"framework":"mindspore","reach":null},{"paper_slug":"auto-deeplab-hierarchical-neural-architecture","repo_url":"https://github.com/2024-MindSpore-1/Code6/tree/main/Auto-DeepLab","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":0,"framework":"mindspore","reach":null},{"paper_slug":"auto-deeplab-hierarchical-neural-architecture","repo_url":"https://github.com/Mind23-2/MindCode-7","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":0,"framework":"mindspore","reach":{"status":"ok"}},{"paper_slug":"auto-deeplab-hierarchical-neural-architecture","repo_url":"https://github.com/MindSpore-paper-code-3/code6/tree/main/Auto-DeepLab","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":0,"framework":"mindspore","reach":null},{"paper_slug":"auto-deeplab-hierarchical-neural-architecture","repo_url":"https://github.com/mindspore-ai/models/tree/master/research/cv/Auto-DeepLab","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":0,"framework":"mindspore","reach":null}],"tasks":[{"task_slug":null,"task_name":"GPU"},{"task_slug":"image-classification","task_name":"Image Classification"},{"task_slug":"image-segmentation","task_name":"Image Segmentation"},{"task_slug":"architecture-search","task_name":"Neural Architecture Search"},{"task_slug":"segmentation","task_name":"Segmentation"},{"task_slug":"semantic-segmentation","task_name":"Semantic Segmentation"},{"task_slug":"image-classification","task_name":"image-classification"}],"methods":[{"method_slug":"lstm","method_name":"LSTM"},{"method_slug":"sigmoid-activation","method_name":"Sigmoid Activation"},{"method_slug":"softmax","method_name":"Softmax"},{"method_slug":"tanh-activation","method_name":"Tanh Activation"}],"datasets_introduced":[],"methods_introduced":[],"results":[{"leaderboard":"/sota/semantic-segmentation-on-ade20k","task":"Semantic Segmentation","dataset":"ADE20K","model":"Auto-DeepLab-L","rank_in_archive_order":205,"of":235,"metrics":{"Validation mIoU":"43.98"},"uses_additional_data":false},{"leaderboard":"/sota/semantic-segmentation-on-ade20k-val","task":"Semantic Segmentation","dataset":"ADE20K val","model":"Auto-DeepLab-L","rank_in_archive_order":87,"of":95,"metrics":{"Pixel Accuracy":"81.72","mIoU":"43.98"},"uses_additional_data":false},{"leaderboard":"/sota/semantic-segmentation-on-cityscapes","task":"Semantic Segmentation","dataset":"Cityscapes test","model":"Auto-DeepLab-L","rank_in_archive_order":31,"of":105,"metrics":{"Mean IoU (class)":"82.1%"},"uses_additional_data":false},{"leaderboard":"/sota/semantic-segmentation-on-cityscapes-val","task":"Semantic Segmentation","dataset":"Cityscapes val","model":"Auto-DeepLab-L","rank_in_archive_order":52,"of":99,"metrics":{"mIoU":"80.33%"},"uses_additional_data":false},{"leaderboard":"/sota/semantic-segmentation-on-pascal-voc-2012","task":"Semantic Segmentation","dataset":"PASCAL VOC 2012 test","model":"Auto-DeepLab-L","rank_in_archive_order":9,"of":51,"metrics":{"Mean IoU":"85.6%"},"uses_additional_data":false},{"leaderboard":"/sota/semantic-segmentation-on-pascal-voc-2012-val","task":"Semantic Segmentation","dataset":"PASCAL VOC 2012 val","model":"Auto-DeepLab-L","rank_in_archive_order":7,"of":29,"metrics":{"mIoU":"82.04%"},"uses_additional_data":false}],"syntology":{"syntology_url":"https://syntology.ai/paper/1901.02985","atlas_url":"https://app.syntology.ai/?focus=1901.02985","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"1901.02985"}},"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/tensorflow/models","reach":null},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/Mind23-2/MindCode-7","reach":{"status":"ok"}},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/NoamRosenberg/autodeeplab","reach":{"status":"gone","observed_at":"2026-09-18","how":"tree_404+repo_404"}},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/2024-MindSpore-1/Code6/tree/main/Auto-DeepLab","reach":null},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/MenghaoGuo/AutoDeeplab","reach":{"status":"ok"}},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/mindspore-ai/models/tree/master/research/cv/Auto-DeepLab","reach":null},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/MindSpore-paper-code-3/code6/tree/main/Auto-DeepLab","reach":null},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/Mind23-2/MindCode-26","reach":{"status":"ok"}},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/2023-MindSpore-4/Code8/tree/main/Auto-DeepLab","reach":null},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/2023-MindSpore-1/ms-code-16/tree/main/Auto-DeepLab","reach":null},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/Dawars/auto_deeplab-pytorch","reach":null},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/zhangximing666/Auto-DeepLab_MS","reach":{"status":"ok","spdx":"MIT"}}],"summary":{"ran_draft_wrong":1,"unverified":3},"by_repo_kind":{"listed":{"samples":4,"ran":1,"repositories":2}},"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":"0cdd8e06cd2a8764","entry":"fixed_padding","repo":"Dawars/auto_deeplab-pytorch","repo_kind":"listed","path":"layers.py","file_url":"https://github.com/Dawars/auto_deeplab-pytorch/blob/HEAD/layers.py","link_basis":"first_harvest_node","language":"python","status":"ran_draft_wrong","verification_level":1,"contract_check":"MISDECLARED","metamorphic_tier":"invariant","behaviour_fingerprint":true,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"0cdd8e06cd2a8764"}},{"code_sha256_prefix":"a0060f94b2278f59","entry":"decode_segmap","repo":"zhangximing666/Auto-DeepLab_MS","repo_kind":"listed","path":"postprocess.py","file_url":"https://github.com/zhangximing666/Auto-DeepLab_MS/blob/HEAD/postprocess.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"a0060f94b2278f59"}},{"code_sha256_prefix":"5d0bd5aa3d2a871f","entry":"encode_segmap","repo":"zhangximing666/Auto-DeepLab_MS","repo_kind":"listed","path":"build_mindrecord.py","file_url":"https://github.com/zhangximing666/Auto-DeepLab_MS/blob/HEAD/build_mindrecord.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"5d0bd5aa3d2a871f"}},{"code_sha256_prefix":"07109e3ece74dcc0","entry":"get_color","repo":"zhangximing666/Auto-DeepLab_MS","repo_kind":"listed","path":"postprocess.py","file_url":"https://github.com/zhangximing666/Auto-DeepLab_MS/blob/HEAD/postprocess.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"07109e3ece74dcc0"}}]},"arxiv_metadata":null,"syntology_extracted_results":null}