List your company
ResearchFormerly WizWand

Anomaly Detection on MVTec AD (image-level AUROC)

99.72AUROC (Image-level)ReConPatch
64.443273.601682.7691.9184May 26, 2023Nov 7, 2023Apr 20, 2024Oct 2, 2024Mar 16, 2025Aug 28, 2025Feb 10, 2026
Updated 8mo ago

Evaluation results

MethodLinks
ReConPatchBackbone=WRN-101 & RNext-101 & DenseN-201, Image size=480×4802023.0599.72
CPR2024.0799.7
ReConPatchBackbone=WRN-101 & RNext-101 & DenseN-201, Image size=320×3202023.0599.67
PNI (w/ refine)Backbone=WRN-101 & RNext-101 & DenseN-201, Image size=480×4802023.0599.63
PatchCoreBackbone=WRN-101 & RNext-101 & DenseN-201, Image size=320×3202023.0599.6
ADClickSupervision=weak2024.0799.6
MemKD2024.0899.6
PRN2024.0799.4
RD++2024.0899.4
DMDD2024.0899.4
BGAD2024.0799.3
THFRRequires Training=Yes, Network=CNN, Structural Capability=Yes, Logical Capability=Yes, FPS=7.692024.0699.2
PatchcoreRequires Training=Yes, Network=CNN, Structural Capability=Yes, Logical Capability=No, FLOPs (Gb)=11.4, FPS=25.12024.0699.1
DeSTSeg2024.0898.6
GLCFRequires Training=Yes, Network=Transformer, Structural Capability=Yes, Logical Capability=Yes, FLOPs (Gb)=52.6, FPS=11.22024.0698.6
RD2024.0898.5
SAM-LADRequires Training=No, Network=Transformer, Structural Capability=Yes, Logical Capability=Yes, FLOPs (Gb)=54.7, FPS=8.92024.0698.4
CRAD2026.0298.2
DPDLNumber of training anomaly examples=102025.0297.7
HLGFA2026.0297.5
AHLNumber of training anomaly examples=102025.0297
DRA2024.0795.9
DRANumber of training anomaly examples=102025.0295.9
STPM2024.0895.5
PadimRequires Training=Yes, Network=CNN, Structural Capability=Yes, Logical Capability=No, FPS=4.62024.0695.5
Pull & Push2024.0894.8
DevNet2024.0794.5
DevNetNumber of training anomaly examples=102025.0294.5
NGAL2026.0294.5
FLOSNumber of training anomaly examples=102025.0293.9
DPDLNumber of training anomaly examples=12025.0292.7
SAOENumber of training anomaly examples=102025.0292.6
RD4AD2026.0292.1
AnomalyCLIP2026.0291.6
MLEPNumber of training anomaly examples=102025.0290.7
AHLNumber of training anomaly examples=12025.0290.1
DRANumber of training anomaly examples=12025.0288.3
US2024.0887.7
SPADERequires Training=Yes, Network=CNN, Structural Capability=Yes, Logical Capability=No, FPS=0.92024.0685.5
SAOENumber of training anomaly examples=12025.0283.4
DevNetNumber of training anomaly examples=12025.0278
FLOSNumber of training anomaly examples=12025.0275.5
MLEPNumber of training anomaly examples=12025.0274.4
AERequires Training=No, Network=CNN, Structural Capability=Yes, Logical Capability=No, FLOPs (Gb)=5.0, FPS=251.12024.0671
f-AnoGANRequires Training=No, Network=CNN, Structural Capability=Yes, Logical Capability=No, FLOPs (Gb)=7.7, FPS=133.42024.0665.8