List your company

Anomaly Detection on MVTec AD 1.0 (test)

100Image AUROCSIA
70.98478.51786.0593.583Jul 4, 2022Feb 15, 2023Sep 30, 2023May 13, 2024Dec 26, 2024Aug 9, 2025Mar 24, 2026

Evaluation results

MethodLinks
SIACategory=Grid2024.03100———————————————————99.5197.28
SIACategory=Leather2024.03100———————————————————99.7696.22
SIACategory=Bottle2024.03100———————————————————99.395.62
SIACategory=Hazelnut2024.03100———————————————————99.6893.14
SIACategory=Transistor2024.03100———————————————————9892.92
DTDCategory=Carpet2024.03100———————————————————99.2796.96
DTDCategory=Grid2024.03100———————————————————99.5797.14
NSACategory=Grid2024.03100———————————————————99.3291.31
CutPasteCategory=Grid2024.03100———————————————————99.1892.53
SIACategory=Tile2024.0399.96———————————————————99.4497.7
DTDCategory=Texture AVG2024.0399.93———————————————————99.2595.4
SIACategory=Carpet2024.0399.84———————————————————99.1996.41
SIACategory=Zipper2024.0399.82———————————————————99.1793.43
SIACategory=Texture AVG2024.0399.8———————————————————99.2295.63
NSACategory=Carpet2024.0399.8———————————————————98.688.77
SIACategory=Metal Nut2024.0399.76———————————————————98.5894.39
SIACategory=Total AVG2024.0399.65———————————————————99.0393.07
CutPasteCategory=Texture AVG2024.0399.61———————————————————97.9988.87
SimpleNet2024.0399.6—————————————————————
RealNetBackbone=WideResNet502024.0399.6—————————————————————
SIACategory=Object AVG2024.0399.57———————————————————98.9391.79
SIACategory=Capsule2024.0399.56———————————————————99.3284.48
DTDCategory=Total AVG2024.0399.52———————————————————98.9291.4
NSACategory=Texture AVG2024.0399.49———————————————————97.5790.11
SIACategory=Toothbrush2024.0399.44———————————————————98.7191.57
FastFlow2024.0399.4—————————————————————
RD++2024.0399.4—————————————————————
NSACategory=Total AVG2024.0399.35———————————————————98.2890.73
DTDCategory=Object AVG2024.0399.31———————————————————98.7589.4
BGADtraining_setting=Multi-class, abnormal_samples=10, supervision=Supervised2022.0799.3—————————————————————
NSACategory=Object AVG2024.0399.28———————————————————98.6391.04
CutPasteCategory=Carpet2024.0399.24———————————————————98.4293.85
SIACategory=Wood2024.0399.21———————————————————98.2290.54
SIACategory=Cable2024.0399.19———————————————————98.193.38
SIACategory=Pill2024.0399.13———————————————————99.0291.04
PatchCore2024.0399.1—————————————————————
SIACategory=Screw2024.0398.83———————————————————99.4587.9
CutPasteCategory=Total AVG2024.0398.73———————————————————98.1486.4
TFA-Net2026.0398.7—99.899.710010099.7—10096.594.810099.196.89810099.896————
DeSTSeg2024.0398.6—————————————————————
PNPT2024.0698.4———————————————————97.3—
CutPasteCategory=Object AVG2024.0398.29———————————————————98.2285.16
DSR2024.0398.2—————————————————————
DRAEM+SSPCAB2024.0398—————————————————————
DraemMethod Paradigm=Reconstruction-based methods2026.0398—9799.910099.699.1—99.291.898.510098.798.993.910093.1100————
DiffAD2024.0397.9—————————————————————
MBPFMMethod Paradigm=Embedding-based methods2026.0397.5—1009810099.699.5—10098.894.510010096.591.888.697.897.4————
DIADLearning Paradigm=Implicit normality learning2024.0697.3———————————————————96—
UniAD2024.0396.6—————————————————————
FCDDtraining_setting=Multi-class, abnormal_samples=10, supervision=Supervised2022.0796.5—————————————————————
UniADLearning Paradigm=Implicit normality learning2024.0696.5———————————————————96.8—
PatchcoreLearning Paradigm=Explicit normality learning2024.0696.5———————————————————96.8—
DRAtraining_setting=Multi-class, abnormal_samples=10, supervision=Supervised2022.0796.1—————————————————————
SimpleNetLearning Paradigm=Implicit normality learning2024.0695.2———————————————————84.8—
IntraMethod Paradigm=Reconstruction-based methods2026.0395—98.910010098.297.5—10070.386.595.796.990.295.710095.899.4————
DevNettraining_setting=Multi-class, abnormal_samples=10, supervision=Supervised2022.0794.8—————————————————————
DFRMethod Paradigm=Reconstruction-based methods2026.0393.8—————————————————————
RD4ADLearning Paradigm=Implicit normality learning2024.0693.4———————————————————96—
GCPFMethod Paradigm=Embedding-based methods2026.0393.1—————————————————————
Pacth SVDDMethod Paradigm=Embedding-based methods2026.0392.1—92.994.690.997.896.5—98.690.376.7929486.181.310091.597.9————
RIADMethod Paradigm=Reconstruction-based methods2026.0391.7—84.299.610098.793—99.981.988.483.388.583.884.510090.998.1————
TrustMAEMethod Paradigm=Reconstruction-based methods2026.0390.9—97.499.195.197.399.8—9785.178.898.576.183.383.496.987.587.5————
DraemLearning Paradigm=Implicit normality learning2024.0688.1———————————————————87.2—
SPADEMethod Paradigm=Embedding-based methods2026.0385.5—92.847.395.496.595.8—97.284.89188.17180.166.788.990.396.6————
PadimLearning Paradigm=Explicit normality learning2024.0685.1———————————————————88.9—
MKDLearning Paradigm=Implicit normality learning2024.0681.9———————————————————84.9—
P-SVDDLearning Paradigm=Explicit normality learning2024.0672.1———————————————————80.7—
AEDistance metric=SSIM2020.12——8794785973—93829497899196829088—86——
AEDistance metric=L22020.12——5990755173—86868895868596938677—82——
AE-l22022.10—82————————————————————
AE-SSIM2022.10—86————————————————————
AnoGAN2020.12——5458645062—86788487768780908078—74——
CFA2022.06———————99.6——————————99.299.3——
CFASupervision=Unsupervised, Backbone=WideResNet50, Resolution=256x2562022.12——99.398.610099.2100—10099.997.410010097.795.110010099.5—99.1——
CFA++2022.06———————99.8——————————99.499.5——
CFLOW2022.06———————98.7——————————9898.3——
CFLOWSupervision=Unsupervised, Backbone=WideResNet50, Resolution=256x2562022.12——97.698.199.997.198.7—99.997.69710098.596.293.198.892.997.1—97.5——
CNN-FD2020.12——7259879391—78798472826887776676—78——
CutPasteImage Size=2562022.01——93.910010094.699.197.598.281.298.298.399.994.988.799.496.199.995.596.1——
CutPaste2022.06———————98.4——————————94.195.5——
CutPasteEvaluation Case=Unified2022.06——93.693.293.488.680.4—67.969.26380.96071.485.263.957.993.5—77.5——
CutPasteEvaluation Case=Separate2022.06——93.910010094.699.1—98.281.298.298.399.994.988.799.496.199.9—96.1——
CutPaste2022.10—97.1————————————————————
DAADImage Size=2562022.01——86.695.786.288.298.29197.684.476.792.175.89098.799.287.685.988.889.5——
DevNetSupervision=Supervised, Resolution=256x2562022.12——82.590.692.299.997.9—99.798.771.999.798.887.197.279.289.199.1—92.2——
DFRRegional feature scales=f{1:12}2020.12——9698998694—95889898909699987596—94——
DFRRegional feature scales=f{1:16}2020.12——9798988793—97929999939799998096—95——
DFR2022.10—95————————————————————
Diff2022.10—94.7————————————————————
DRASupervision=Supervised, Resolution=256x2562022.12——92.598.698.910099.1—10094.295.110099.188.399.587.588.399.7—96.1——
DRAEM2022.10——9799.910099.699.1—99.291.898.510098.798.993.910093.1100—98——
DRAEMEvaluation Case=Unified2022.06——9899.398.799.899.8—97.557.865.393.772.882.29290.674.898.8—88.1——
DRAEMEvaluation Case=Separate2022.06——9799.910099.699.1—99.291.898.510098.798.993.910093.1100—98——
DRAEMSupervision=Unsupervised, Backbone=reconstructive encoder, Resolution=256x2562022.12——96.999.910010099.5—9890.991.310010097.198.710091.7100—97.6——
EdgRec2022.10—97.8————————————————————
GANomaly (GN)Image Size=2562022.01——69.970.884.279.483.477.589.275.773.278.57074.374.665.379.274.575.576.2——
GTImage Size=2562022.01——43.761.984.141.761.158.574.478.36735.981.3635097.286.98271.667.2——
Intra2022.10—95.9————————————————————
ITAE2022.10—83.9————————————————————
KDADSupervision=Unsupervised, Resolution=256x2562022.12——80.375.392.391.594.5—99.290.381.498.877.184.482.497.184.993.7—88.2——