{"about":{"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.","site":"https://codewithpapers.app","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","syntology":{"site":"https://syntology.ai","developers":"https://syntology.ai/developers","mcp":{"server":"https://syntology.ai/mcp","transport":"streamable-http","server_card":"https://syntology.ai/.well-known/mcp/server-card.json","auth":{"type":"trial token, no account","trial_token":"https://syntology.ai/api/oauth/trial/token","method":"POST","docs":"https://syntology.ai/developers"}},"have":"https://syntology.ai/api/graph/have?x=<method, arXiv id or title> (free, answers coverage only)","paper_base":"https://syntology.ai/paper/","atlas_base":"https://app.syntology.ai/?focus="},"machine_readable":[{"url":"https://codewithpapers.app/llms.txt","what":"the machine catalog: every machine-readable file, counted"},{"url":"https://codewithpapers.app/index/manifest.json","what":"paper-to-code index by arXiv id, with Syntology's counts"},{"url":"https://codewithpapers.app/search/manifest.json","what":"site search index (titles, authors) and its files"},{"url":"https://codewithpapers.app/download","what":"bulk files: Syntology's layer, described there"},{"url":"https://codewithpapers.app/build_manifest.json","what":"the build record: inputs, counts, exclusions, probes"}]},"url":"/paper/asymmetric-student-teacher-networks-for","title":"Asymmetric Student-Teacher Networks for Industrial Anomaly Detection","arxiv_id":"2210.07829","date":"2022-10-14","proceeding":null,"authors":["Marco Rudolph","Tom Wehrbein","Bodo Rosenhahn","Bastian Wandt"],"abstract":"Industrial defect detection is commonly addressed with anomaly detection (AD) methods where no or only incomplete data of potentially occurring defects is available. This work discovers previously unknown problems of student-teacher approaches for AD and proposes a solution, where two neural networks are trained to produce the same output for the defect-free training examples. The core assumption of student-teacher networks is that the distance between the outputs of both networks is larger for anomalies since they are absent in training. However, previous methods suffer from the similarity of student and teacher architecture, such that the distance is undesirably small for anomalies. For this reason, we propose asymmetric student-teacher networks (AST). We train a normalizing flow for density estimation as a teacher and a conventional feed-forward network as a student to trigger large distances for anomalies: The bijectivity of the normalizing flow enforces a divergence of teacher outputs for anomalies compared to normal data. Outside the training distribution the student cannot imitate this divergence due to its fundamentally different architecture. Our AST network compensates for wrongly estimated likelihoods by a normalizing flow, which was alternatively used for anomaly detection in previous work. We show that our method produces state-of-the-art results on the two currently most relevant defect detection datasets MVTec AD and MVTec 3D-AD regarding image-level anomaly detection on RGB and 3D data.","url_abs":"https://arxiv.org/abs/2210.07829v2","url_pdf":"https://arxiv.org/pdf/2210.07829v2.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":"asymmetric-student-teacher-networks-for","repo_url":"https://github.com/marco-rudolph/ast","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok"}}],"tasks":[{"task_slug":"3d-anomaly-detection","task_name":"3D Anomaly Detection"},{"task_slug":"anomaly-detection","task_name":"Anomaly Detection"},{"task_slug":"defect-detection","task_name":"Defect Detection"},{"task_slug":"density-estimation","task_name":"Density Estimation"},{"task_slug":"rgb-3d-anomaly-detection-and-segmentation","task_name":"RGB+3D Anomaly Detection and Segmentation"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[{"leaderboard":"/sota/anomaly-detection-on-mvtec-3d-ad-1","task":"Anomaly Detection","dataset":"MVTEC 3D-AD","model":"AST","rank_in_archive_order":2,"of":2,"metrics":{"Detection AUROC":"93.7","Segmentation AUROC":"97.6"},"uses_additional_data":true},{"leaderboard":"/sota/anomaly-detection-on-mvtec-ad","task":"Anomaly Detection","dataset":"MVTec AD","model":"AST","rank_in_archive_order":44,"of":148,"metrics":{"Detection AUROC":"99.2","Segmentation AUROC":"95.0"},"uses_additional_data":false},{"leaderboard":"/sota/anomaly-detection-on-mvtec-loco-ad","task":"Anomaly Detection","dataset":"MVTec LOCO AD","model":"AST","rank_in_archive_order":40,"of":40,"metrics":{"Detection AUROC (only logical)":"79.7","Detection AUROC (only structural)":"87.1","Segmentation AU-sPRO (until FPR 5%)":"42.7"},"uses_additional_data":false},{"leaderboard":"/sota/anomaly-detection-on-visa","task":"Anomaly Detection","dataset":"VisA","model":"AST","rank_in_archive_order":21,"of":50,"metrics":{"Detection AUROC":"94.9","Segmentation AUPRO (until 30% FPR)":"81.5"},"uses_additional_data":false},{"leaderboard":"/sota/rgb-3d-anomaly-detection-and-segmentation-on","task":"RGB+3D Anomaly Detection and Segmentation","dataset":"MVTEC 3D-AD","model":"AST","rank_in_archive_order":5,"of":9,"metrics":{"Detection AUCROC":"0.937","Segmentation AUCROC":"0.976"},"uses_additional_data":false}],"syntology":{"syntology_url":"https://syntology.ai/paper/2210.07829","atlas_url":"https://app.syntology.ai/?focus=2210.07829","mcp":null,"developers":"https://syntology.ai/developers","held":true,"reason":"arxiv_id_in_atlas","checked_against":{"atlas_ids":549449,"atlas_ids_sha256":"afd72399dafd9906e9062927f70655952385a3d75488558e1f0484e950ab09b7","read_at":"2026-10-05T01:46:21Z"},"agent_calls":[{"tool":"get_citation_path","arguments":{"paper_1":"2210.07829"},"arguments_in":{"paper_2":"another paper's arXiv id or title"},"call":"get_citation_path(paper_1=\"2210.07829\", paper_2=\"…\")"},{"tool":"get_concepts_for_paper","arguments":{"arxiv_id":"2210.07829"},"call":"get_concepts_for_paper(arxiv_id=\"2210.07829\")"}],"by_repo":[{"repo":"marco-rudolph/ast","kind":"official","samples":0,"ran":0,"instrument":0,"unverified":0}],"by_repo_is":"one row per repository in the archive's code links, Syntology's graph links or the samples; samples 0 means no sample is linked to this paper, a statement about Syntology's coverage, not about the repository: with harvested_for_other_papers present, Syntology harvested the repository for that many other papers, and without it, Syntology harvested nothing from it; placed_by_identical_code counts samples whose link records no repository, placed here because identical code was harvested from this repository for another paper; ran counts executions on a synthesized input and instrument counts failures of Syntology's instrument, not of the code"},"code_links_note":null,"arxiv_metadata":null,"syntology_extracted_results":null,"record_sha256":"000f2c00d2a22c823681a3968ca27c6300bd86f03c37619a99d8cb0fe28cf571","record_changed_at":"2026-09-28","record_changed_at_basis":"first_hashed"}