Tasks SotA Datasets Papers Methods Submit About

Papers With Code 2

A community resource for machine learning research: papers, code, benchmarks, and state-of-the-art results.

Explore

Notable Benchmarks All SotA Datasets Papers Methods

Community

Submit Results About

Data sourced from the PWC Archive (CC-BY-SA 4.0). Built by the community, for the community.

Models/DRAEM

DRAEM

Reported on 17 benchmarks across 3 tasks · 1 paper · 1 SOTA

Note: results are matched by exact model name. Different papers may use the same name for different model variants.

Methodology15 results

Anomaly DetectiononMVTec AD
Segmentation AP· uses extra data· 2021-08-17
68.4
best: 87.6 (WeakREST-Block)
SOTA
DRAEM -- A discriminatively trained reconstruction embedding for surface anomaly detection arXiv:2108.07610
Anomaly DetectiononMVTec AD
Detection AUROC· uses extra data· 2021-08-17
98
best: 99.9 (GLASS)
DRAEM -- A discriminatively trained reconstruction embedding for surface anomaly detection arXiv:2108.07610
Anomaly DetectiononMVTec AD
Segmentation AUROC· uses extra data· 2021-08-17
97.3
best: 99.7 (WeakREST-Block)
DRAEM -- A discriminatively trained reconstruction embedding for surface anomaly detection arXiv:2108.07610
Anomaly DetectiononVisA
Segmentation AUPRO (until 30% FPR)· 2021-08-17
73.1
best: 96.1 (AnomalyDINO-S (full-shot))
DRAEM -- A discriminatively trained reconstruction embedding for surface anomaly detection arXiv:2108.07610
Anomaly DetectiononMVTec LOCO AD
Avg. Detection AUROC· 2021-08-17
73.6
best: 95.3 (CSAD)
DRAEM -- A discriminatively trained reconstruction embedding for surface anomaly detection arXiv:2108.07610
Anomaly DetectiononMVTec LOCO AD
Detection AUROC (only logical)· 2021-08-17
72.8
best: 98.1 (PSAD)
DRAEM -- A discriminatively trained reconstruction embedding for surface anomaly detection arXiv:2108.07610
Anomaly DetectiononMVTec LOCO AD
Detection AUROC (only structural)· 2021-08-17
74.4
best: 95.9 (PUAD-M)
DRAEM -- A discriminatively trained reconstruction embedding for surface anomaly detection arXiv:2108.07610
Anomaly DetectiononMVTec LOCO AD
Segmentation AU-sPRO (until FPR 5%)· 2021-08-17
42.6
best: 83.2 (SAM-LAD)
DRAEM -- A discriminatively trained reconstruction embedding for surface anomaly detection arXiv:2108.07610
Anomaly DetectiononGoodsAD
AUPR· 2021-08-17
71
best: 86.1 (PatchCore-100%)
DRAEM -- A discriminatively trained reconstruction embedding for surface anomaly detection arXiv:2108.07610
Anomaly DetectiononGoodsAD
AUROC· 2021-08-17
65.9
best: 86.1 (MiniMaxAD-fr)
DRAEM -- A discriminatively trained reconstruction embedding for surface anomaly detection arXiv:2108.07610
2D ClassificationonGoodsAD
AUPR· 2021-08-17
71
best: 86.1 (PatchCore-100%)
DRAEM -- A discriminatively trained reconstruction embedding for surface anomaly detection arXiv:2108.07610
2D ClassificationonGoodsAD
AUROC· 2021-08-17
65.9
best: 86.1 (MiniMaxAD-fr)
DRAEM -- A discriminatively trained reconstruction embedding for surface anomaly detection arXiv:2108.07610
Anomaly DetectiononAeBAD-V
Detection AUROC
68.1
best: 78.2 (MMR)
Anomaly DetectiononAeBAD-S
Detection AUROC
62.5
best: 87.1 (MSFR)
Anomaly DetectiononAeBAD-S
Segmentation AUPRO
63.6
best: 90.4 (MSFR)

Computer Vision2 results

Anomaly ClassificationonGoodsAD
AUPR· 2021-08-17
71
best: 86.1 (PatchCore-100%)
DRAEM -- A discriminatively trained reconstruction embedding for surface anomaly detection arXiv:2108.07610
Anomaly ClassificationonGoodsAD
AUROC· 2021-08-17
65.9
best: 86.1 (MiniMaxAD-fr)
DRAEM -- A discriminatively trained reconstruction embedding for surface anomaly detection arXiv:2108.07610