Anomaly Detection on DAGM benchmark leaderboard · SOTA2 Research

Anomaly Detection on DAGM

100Average ScoreBo et al.

Evaluation Results

Method	Links
Bo et al.2022.03		100	100	100	100	100	99.9	100	100	100	100	100
MS2023.12		100	—	—	—	—	—	—	—	—	—	—
POUTA2023.12		99.7	100	100	100	98.9	99.9	99	100	100	97.4	100
CCNN2023.12		99.6	99	100	98	99	100	100	100	100	100	100
OCR-GAN2022.03		99.3	99.1	100	99.1	99	100	97.5	99.8	99.8	99.5	99.2
Lin et al.2022.03		99.3	100	94	100	100	100	100	100	99	100	100
DBPI2023.12		99	—	—	—	—	—	—	—	—	—	—
Draem2022.03		98	96.1	98.3	99.5	99.6	92.1	100	99.7	99.9	98.9	96
DRÆM2023.12		96	—	—	—	—	—	—	—	—	—	—
PatchCore2023.12		92.7	87.3	100	93.5	100	91.4	97.9	99.8	61.4	95.4	100
DifferNet2022.03		74.6	59.7	82.9	69.8	97.3	61.2	97	68.5	52.1	78.2	79.1
CutPaste2022.03		66	56.1	87.8	57.1	71.3	47.4	68.8	96.5	53.4	51.9	74.7
Puzzle AE2022.03		59.3	50.7	50.5	58.7	70	63.6	92.3	54	49.1	54.6	49.6
skipGAN2022.03		55.8	58.3	56.1	55.1	53.7	57.4	66.8	52.4	53.7	52.3	52.2