Papers › Beyond Dents and Scratches: Logical Constraints in Unsupervised Anomaly Detection and...
Beyond Dents and Scratches: Logical Constraints in Unsupervised Anomaly Detection and Localization
Paul Bergmann, Kilian Batzner, Michael Fauser, David Sattlegger, Carsten Steger
The unsupervised detection and localization of anomalies in natural images is an intriguing and challenging problem. Anomalies manifest themselves in very different ways and an ideal benchmark dataset for this task should contain representative examples for all of them. We find that existing datasets are biased towards local structural anomalies such as scratches, dents, or contaminations. In particular, they lack anomalies in the form of violations of logical constraints, e.g., permissible objects occurring in invalid locations. We contribute a new dataset based on industrial inspection scenarios that evenly covers both types of anomalies. We provide pixel-precise ground truth data for each anomalous region and define a generalized evaluation metric that addresses localization ambiguities that can arise for logical anomalies. Furthermore, we propose a novel algorithm that improves over the state of the art in the joint detection of structural and logical anomalies. It consists of a local and a global network branch. The first one inspects confined regions independent of their spatial locations in the input image and is primarily responsible for the detection of entirely new local structures. The second one learns a globally consistent representation of the training data through a bottleneck that enables the detection of violations of long-range dependencies, a key characteristic of many logical anomalies. We perform extensive evaluations on our new dataset to corroborate our claims.
Code
No code repository is listed for this paper in the archive or in Syntology's graph.
Code Syntology ran Syntology
Not run by Syntology. Nothing on this page verifies that the listed code works.
Tasks
Results from the paper archive 2025-07-28
| Task | Dataset | Model | Metric | Value | Rank at snapshot | Leaderboard | Report |
|---|---|---|---|---|---|---|---|
| Anomaly Detection | MVTec LOCO AD | GCAD | Avg. Detection AUROC | 83.3 | #22 of 40 | Archive leaderboard | report |
| Anomaly Detection | MVTec LOCO AD | GCAD | Detection AUROC (only logical) | 86.0 | #22 of 40 | Archive leaderboard | report |
| Anomaly Detection | MVTec LOCO AD | GCAD | Detection AUROC (only structural) | 80.6 | #22 of 40 | Archive leaderboard | report |
| Anomaly Detection | MVTec LOCO AD | GCAD | Segmentation AU-sPRO (until FPR 5%) | 70.1 | #22 of 40 | Archive leaderboard | report |
| Anomaly Detection | MVTec LOCO AD | Variation Model | Avg. Detection AUROC | 57.7 | #37 of 40 | Archive leaderboard | report |
| Anomaly Detection | MVTec LOCO AD | Variation Model | Detection AUROC (only logical) | 56.5 | #37 of 40 | Archive leaderboard | report |
| Anomaly Detection | MVTec LOCO AD | Variation Model | Detection AUROC (only structural) | 58.9 | #37 of 40 | Archive leaderboard | report |
| Anomaly Detection | MVTec LOCO AD | Variation Model | Segmentation AU-sPRO (until FPR 5%) | 22.5 | #37 of 40 | Archive leaderboard | report |
| Anomaly Detection | MVTec LOCO AD | L2AE | Avg. Detection AUROC | 57.3 | #38 of 40 | Archive leaderboard | report |
| Anomaly Detection | MVTec LOCO AD | L2AE | Detection AUROC (only logical) | 58.1 | #38 of 40 | Archive leaderboard | report |
| Anomaly Detection | MVTec LOCO AD | L2AE | Detection AUROC (only structural) | 56.5 | #38 of 40 | Archive leaderboard | report |
| Anomaly Detection | MVTec LOCO AD | L2AE | Segmentation AU-sPRO (until FPR 5%) | 37.8 | #38 of 40 | Archive leaderboard | report |
| Anomaly Detection | VisA | GCAD | Detection AUROC | 89.1 | #30 of 50 | Archive leaderboard | report |
| Anomaly Detection | VisA | GCAD | Segmentation AUPRO (until 30% FPR) | 83.7 | #30 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