Research map / Detection and segmentation / Open-set and few-shot detection
The MVTec AD 2 Dataset: Advanced Scenarios for Unsupervised Anomaly Detection
Lars Heckler-Kram, Jan-Hendrik Neudeck, Ulla Scheler, Rebecca König, Carsten T. Steger
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| change lighting from regular to mixed conditions in the test set | MSFlow · MVTec AD 2 (mean of 8 objects) | AU-PRO 0.05, %: 24.3 -> 11.9 | Hurt |
| change lighting from regular to mixed conditions in the test set | RD · MVTec AD 2 (mean of 8 objects) | AU-PRO 0.05, %: 26.4 -> 25.0 | Hurt |
| change lighting from regular to mixed conditions in the test set | EfficientAD · MVTec AD 2 (mean of 8 objects) | AU-PRO 0.05, %: 30.8 -> 19.2 | Hurt |
| move from established benchmark MVTec AD to MVTec AD 2 | EfficientAD · segmentation AU-PRO 0.3 | AU-PRO 0.3, %: 93.5 -> 58.7 | Hurt |
| increase input image size from 256x256 up to half original resolution | PatchCore and MSFlow · MVTec AD 2 Rice | AU-PRO 0.05: approximately double for the largest image size | Improved |
| increase input image size from 256x256 up to half original resolution | PatchCore · MVTec AD 2 Rice | inference time: 2 seconds per image | Hurt |
| apply threshold of validation mean plus three standard deviations (F1) instead of threshold-free AU-PRO | PatchCore · MVTec AD 2 TESTprivate | pixel F1 vs AU-PRO 0.05, %: AU-PRO 28.8 vs F1 3.7 | Mixed |
Closest papers by meaning
- Supervised Anomaly Detection for Complex Industrial Images (CVPR 2024 · similarity 0.88)
- CableInspect-AD: An Expert-Annotated Anomaly Detection Dataset (NeurIPS 2024 · similarity 0.87)
- MuSc: Zero-Shot Industrial Anomaly Classification and Segmentation with Mutual Scoring of the Unlabeled Images (ICLR 2024 · similarity 0.87)
- Real-IAD: A Real-World Multi-View Dataset for Benchmarking Versatile Industrial Anomaly Detection (CVPR 2024 · similarity 0.86)
- AnomalyNCD: Towards Novel Anomaly Class Discovery in Industrial Scenarios (CVPR 2025 · similarity 0.86)
- 3CAD: A Large-Scale Real-World 3C Product Dataset for Unsupervised Anomaly Detection (AAAI 2025 · similarity 0.86)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
