Research map / Detection and segmentation / CNN features
Glancing at the Patch: Anomaly Localization With Global and Local Feature Comparison
Shenzhi Wang, Liwei Wu, Lei Cui, Yujun Shen
Check 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 |
|---|---|---|---|
| compare each patch's local feature with a feature predicted from its surrounding context | Ours · MVTec AD | PRO: 0.914 -> 0.961 | Improved |
| use global-local feature comparison for localization | Ours · MVTec AD | pixel-level AUROC: 0.89 -> 0.91 | Improved |
| use global-local feature comparison for anomaly detection | Ours · CIFAR-10 | image-level AUROC: 0.664 -> 0.705 | Improved |
| combine IAD and DAD scores with λs=0.8 rather than use only DAD (ablation) | Ours · MVTec AD | PRO: 0.912 -> 0.961 | Improved |
| combine IAD and DAD scores with λs=0.8 rather than use only IAD (ablation) | Ours · MVTec AD | PRO: 0.954 -> 0.961 | Improved |
| increase the IAD score weight from λs=0.0 to λs=0.8 (ablation) | Ours · MVTec AD | PRO: 0.912 -> 0.961 | Improved |
| increase the IAD score weight from λs=0.8 to λs=1.0 (ablation) | Ours · MVTec AD | PRO: 0.961 -> 0.954 | Hurt |
Closest papers by meaning
- SimpleNet: A Simple Network for Image Anomaly Detection and Localization (CVPR 2023 · similarity 0.90)
- MetaUAS: Universal Anomaly Segmentation with One-Prompt Meta-Learning (NeurIPS 2024 · similarity 0.88)
- Correcting Deviations from Normality: A Reformulated Diffusion Model for Multi-Class Unsupervised Anomaly Detection (CVPR 2025 · similarity 0.87)
- Boosting Fine-Grained Visual Anomaly Detection with Coarse-Knowledge-Aware Adversarial Learning (AAAI 2025 · similarity 0.87)
- UniVAD: A Training-free Unified Model for Few-shot Visual Anomaly Detection (CVPR 2025 · similarity 0.87)
- PatchGuard: Adversarially Robust Anomaly Detection and Localization through Vision Transformers and Pseudo Anomalies (CVPR 2025 · similarity 0.87)
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.
