Papers › Asymmetric Student-Teacher Networks for Industrial Anomaly Detection

Asymmetric Student-Teacher Networks for Industrial Anomaly Detection

14 Oct 2022arXiv:2210.07829archive 2025-07-28

Marco Rudolph, Tom Wehrbein, Bodo Rosenhahn, Bastian Wandt

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.

Paper PDF Code

In Syntology Open this paper in Syntology's Atlas, the map of the papers in Syntology's graph and their citations.

Code

marco-rudolph/ast officialmentioned in papermentioned on GitHubpytorch report

Repository list and official/mentioned flags are the archive's, frozen 2025-07-28. Reachability, where shown, is from one Syntology probe window (2026-09-16 to 2026-09-18); repositories not probed show nothing. GitHub stars are not tracked.

Code Syntology ran Syntology

Not run by Syntology. Nothing on this page verifies that the listed code works.

Tasks

3D Anomaly Detection Anomaly Detection Defect Detection Density Estimation RGB+3D Anomaly Detection and Segmentation

Results from the paper archive 2025-07-28

Task	Dataset	Model	Metric	Value	Rank at snapshot	Leaderboard	Report
Anomaly Detection	MVTEC 3D-AD	AST	Detection AUROC	93.7	#2 of 2	Archive leaderboard	report
Anomaly Detection	MVTEC 3D-AD	AST	Segmentation AUROC	97.6	#2 of 2	Archive leaderboard	report
Anomaly Detection	MVTec AD	AST	Detection AUROC	99.2	#44 of 148	Archive leaderboard	report
Anomaly Detection	MVTec AD	AST	Segmentation AUROC	95.0	#44 of 148	Archive leaderboard	report
Anomaly Detection	MVTec LOCO AD	AST	Detection AUROC (only logical)	79.7	#40 of 40	Archive leaderboard	report
Anomaly Detection	MVTec LOCO AD	AST	Detection AUROC (only structural)	87.1	#40 of 40	Archive leaderboard	report
Anomaly Detection	MVTec LOCO AD	AST	Segmentation AU-sPRO (until FPR 5%)	42.7	#40 of 40	Archive leaderboard	report
Anomaly Detection	VisA	AST	Detection AUROC	94.9	#21 of 50	Archive leaderboard	report
Anomaly Detection	VisA	AST	Segmentation AUPRO (until 30% FPR)	81.5	#21 of 50	Archive leaderboard	report
RGB+3D Anomaly Detection and Segmentation	MVTEC 3D-AD	AST	Detection AUCROC	0.937	#5 of 9	Archive leaderboard	report
RGB+3D Anomaly Detection and Segmentation	MVTEC 3D-AD	AST	Segmentation AUCROC	0.976	#5 of 9	Archive leaderboard	report

Ranks are positions in the archive's leaderboards as they stood at the 2025-07-28 snapshot. Results published since then are not among these rows, so a rank here is not a current standing.

Report a problem or propose a change · a person checks every report against the paper or source before anything changes; decisions are listed on /corrections