Papers › Student-Teacher Feature Pyramid Matching for Anomaly Detection

Student-Teacher Feature Pyramid Matching for Anomaly Detection

7 Mar 2021arXiv:2103.04257archive 2025-07-28

Guodong Wang, Shumin Han, Errui Ding, Di Huang

Anomaly detection is a challenging task and usually formulated as an one-class learning problem for the unexpectedness of anomalies. This paper proposes a simple yet powerful approach to this issue, which is implemented in the student-teacher framework for its advantages but substantially extends it in terms of both accuracy and efficiency. Given a strong model pre-trained on image classification as the teacher, we distill the knowledge into a single student network with the identical architecture to learn the distribution of anomaly-free images and this one-step transfer preserves the crucial clues as much as possible. Moreover, we integrate the multi-scale feature matching strategy into the framework, and this hierarchical feature matching enables the student network to receive a mixture of multi-level knowledge from the feature pyramid under better supervision, thus allowing to detect anomalies of various sizes. The difference between feature pyramids generated by the two networks serves as a scoring function indicating the probability of anomaly occurring. Due to such operations, our approach achieves accurate and fast pixel-level anomaly detection. Very competitive results are delivered on the MVTec anomaly detection dataset, superior to the state of the art ones.

PaperPDFCode

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

Code

CuberrChen/STFPM-Paddle mentioned on GitHubpaddle report
Rthete/STPM-mindspore mentioned on GitHubmindspore report
SimonThomine/DistillationAD mentioned on GitHubpytorch report
SimonThomine/RememberingNormality mentioned on GitHubpytorch report
gdwang08/STFPM mentioned on GitHubpytorch report
hcw-00/STPM_anomaly_detection mentioned on GitHubpytorch report
open-edge-platform/geti mentioned on GitHubpytorchApache-2.0 report
openvinotoolkit/anomalib pytorchApache-2.0 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

Anomaly DetectionImage ClassificationUnsupervised Anomaly Detectionimage-classification

Results from the paper archive 2025-07-28

TaskDatasetModelMetricValueRank at snapshotLeaderboardReport
Anomaly Detection MVTec AD STPM Detection AUROC 95.5 #88 of 148 Archive leaderboard report
Anomaly Detection MVTec AD STPM Segmentation AUROC 97.0 #88 of 148 Archive leaderboard report
Anomaly Detection VisA STPM Detection AUROC 83.3 #37 of 50 Archive leaderboard report
Anomaly Detection VisA STPM Segmentation AUPRO (until 30% FPR) 62.0 #37 of 50 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