ML Research Wiki / Benchmarks / Anomaly Detection / MVTec LOCO AD

MVTec LOCO AD

Anomaly Detection Benchmark

Performance Over Time

📊 Showing 35 results | 📏 Metric: Avg. Detection AUROC

Top Performing Models

Rank Model Paper Avg. Detection AUROC Date Code
1 CSAD CSAD: Unsupervised Component Segmentation for Logical Anomaly Detection 95.30 2024-08-28 📦 Tokichan/CSAD
2 PSAD Few Shot Part Segmentation Reveals Compositional Logic for Industrial Anomaly Detection 94.90 2023-12-21 📦 oopil/PSAD_logical_anomaly_detection
3 PUAD-M PUAD: Frustratingly Simple Method for Robust Anomaly Detection 94.40 2024-02-23 📦 LeapMind/PUAD
4 SINBAD+EfficientAD Set Features for Anomaly Detection 94.20 2023-11-24 📦 NivC/SINBAD
5 PUAD-S PUAD: Frustratingly Simple Method for Robust Anomaly Detection 93.10 2024-02-23 📦 LeapMind/PUAD
6 EfficientAD-M EfficientAD: Accurate Visual Anomaly Detection at Millisecond-Level Latencies 90.70 2023-03-25 📦 openvinotoolkit/anomalib 📦 nelson1425/EfficientAD 📦 rximg/EfficientAD
7 SAM-LAD SAM-LAD: Segment Anything Model Meets Zero-Shot Logic Anomaly Detection 90.70 2024-06-02 -
8 SLSG SLSG: Industrial Image Anomaly Detection by Learning Better Feature Embeddings and One-Class Classification 90.30 2023-04-30 -
9 ComAD+PatchCore Component-aware anomaly detection framework for adjustable and logical industrial visual inspection 90.10 2023-05-15 📦 liutongkun/comad
10 EfficientAD-S EfficientAD: Accurate Visual Anomaly Detection at Millisecond-Level Latencies 90.00 2023-03-25 📦 openvinotoolkit/anomalib 📦 nelson1425/EfficientAD 📦 rximg/EfficientAD

All Papers (35)