Research map / Detection and segmentation / Open-set and few-shot detection
MVTec AD -- A Comprehensive Real-World Dataset for Unsupervised Anomaly Detection
Check your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (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 |
|---|---|---|---|
| Use the evaluated methods on object categories rather than relying on one method across all categories | unsupervised anomaly segmentation on MVTec AD · MVTec AD objects | relative per-region overlap | Mixed |
| Use a CNN feature dictionary for texture anomaly detection | CNN Feature Dictionary · MVTec AD textures | ROC AUC | Helped |
| Use the GMM-based Texture Inspection Model instead of other evaluated methods | MVTec AD texture categories · Grid texture | anomaly segmentation | Hurt |
| Use a grayscale-only texture inspection method for color-based defects | GMM-based Texture Inspection Model · MVTec AD textures with color-based defects | anomaly detection | Hurt |
| Apply CNN feature descriptors without spatial-location information to object categories | CNN Feature Dictionary · MVTec AD objects | anomaly detection | Hurt |
| Model varied object shapes and orientations with AnoGAN | AnoGAN · MVTec AD object categories with shape or orientation variation | anomaly detection | Hurt |
| Estimate the segmentation threshold using anomaly-free validation images only | evaluated anomaly detection methods · MVTec AD | relative per-region overlap | Mixed |
Closest papers by meaning
- A Diffusion-Based Framework for Multi-Class Anomaly Detection (AAAI 2024 · similarity 0.87)
- AnomalyNCD: Towards Novel Anomaly Class Discovery in Industrial Scenarios (CVPR 2025 · similarity 0.87)
- Long-Tailed Anomaly Detection with Learnable Class Names (CVPR 2024 · similarity 0.87)
- ADPretrain: Advancing Industrial Anomaly Detection via Anomaly Representation Pretraining (NeurIPS 2025 · similarity 0.86)
- Unseen Visual Anomaly Generation (CVPR 2025 · similarity 0.86)
- UniVAD: A Training-free Unified Model for Few-shot Visual Anomaly Detection (CVPR 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.