{"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/adversarial-complementary-learning-for-weakly","title":"Adversarial Complementary Learning for Weakly Supervised Object Localization","arxiv_id":"1804.06962","date":"2018-04-19","proceeding":"CVPR 2018 6","authors":["Xiaolin Zhang","Yunchao Wei","Jiashi Feng","Yi Yang","Thomas Huang"],"abstract":"In this work, we propose Adversarial Complementary Learning (ACoL) to\nautomatically localize integral objects of semantic interest with weak\nsupervision. We first mathematically prove that class localization maps can be\nobtained by directly selecting the class-specific feature maps of the last\nconvolutional layer, which paves a simple way to identify object regions. We\nthen present a simple network architecture including two parallel-classifiers\nfor object localization. Specifically, we leverage one classification branch to\ndynamically localize some discriminative object regions during the forward\npass. Although it is usually responsive to sparse parts of the target objects,\nthis classifier can drive the counterpart classifier to discover new and\ncomplementary object regions by erasing its discovered regions from the feature\nmaps. With such an adversarial learning, the two parallel-classifiers are\nforced to leverage complementary object regions for classification and can\nfinally generate integral object localization together. The merits of ACoL are\nmainly two-fold: 1) it can be trained in an end-to-end manner; 2) dynamically\nerasing enables the counterpart classifier to discover complementary object\nregions more effectively. We demonstrate the superiority of our ACoL approach\nin a variety of experiments. In particular, the Top-1 localization error rate\non the ILSVRC dataset is 45.14%, which is the new state-of-the-art.","url_abs":"http://arxiv.org/abs/1804.06962v1","url_pdf":"http://arxiv.org/pdf/1804.06962v1.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":"adversarial-complementary-learning-for-weakly","repo_url":"https://github.com/Hayashi-Yudai/ML_models","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"tf","reach":null},{"paper_slug":"adversarial-complementary-learning-for-weakly","repo_url":"https://github.com/junkwhinger/adversarial_complementary_learning","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":null}],"tasks":[{"task_slug":"classification","task_name":"General Classification"},{"task_slug":"object","task_name":"Object"},{"task_slug":"object-localization","task_name":"Object Localization"},{"task_slug":"weakly-supervised-object-localization","task_name":"Weakly-Supervised Object Localization"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[{"leaderboard":"/sota/weakly-supervised-object-localization-on-1","task":"Weakly-Supervised Object Localization","dataset":"ILSVRC 2016","model":"GoogLeNet-ACoL","rank_in_archive_order":2,"of":4,"metrics":{"Top-5 Error":"42.58"},"uses_additional_data":false}],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1804.06962","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"1804.06962"}},"developers":"https://syntology.ai/developers","read_at":"2026-09-24T18:15:14+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/Hayashi-Yudai/ML_models","reach":null},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/junkwhinger/adversarial_complementary_learning","reach":null}],"summary":{"ran_draft_wrong":1},"by_repo_kind":{"listed":{"samples":1,"ran":1,"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":"f4ce937469e77659","entry":"restCrossEtropyLoss","repo":"junkwhinger/adversarial_complementary_learning","repo_kind":"listed","path":"train_acol.py","file_url":"https://github.com/junkwhinger/adversarial_complementary_learning/blob/HEAD/train_acol.py","link_basis":"first_harvest_node","language":"python","status":"ran_draft_wrong","verification_level":1,"contract_check":"MISDECLARED","metamorphic_tier":"invariant","behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"f4ce937469e77659"}}]},"arxiv_metadata":null,"syntology_extracted_results":null}