Anomaly Detection on MVTec 3D-AD (AUPRO@30% per class) benchmark leaderboard · SOTA2 Research

Anomaly Detection on MVTec 3D-AD (AUPRO@30% per class)

98.4AUPRO@30% (Bagel)MTSJM

Evaluation Results

Method	Links
MTSJM2026.03		98.4	98.1	98.3	96.8	93.9	95.1	97.7	98.5	97.4	97.3	97.2
CMDR–IAD2026.03		98	97.3	98.2	95.9	96.6	97.1	98.2	98.2	98	98.2	97.6
BTM2026.03		97.9	97.2	98	97.6	97.7	90.5	97.8	98.2	96.8	97.5	96.9
CFM2026.03		97.9	97.2	98.2	94.5	95	96.8	98	98.2	97.5	98.1	97.1
MTSJM2026.03		97.6	96.9	97.5	91.4	94.9	86.9	97.5	96.3	97	96.6	95.3
BTF2026.03		97.6	96.9	97.9	97.3	93.3	88.8	97.5	98.1	95	97.1	95.9
AST2026.03		97	94.7	98.1	93.6	91.2	91.8	97.9	98.3	88.7	94.2	94.6
M3DM2026.03		97	97.1	97.9	95	94.1	93.2	97.7	97.1	97.1	97.5	96.4
CMDIAD2026.03		97	97.1	97.7	93.2	93.4	94.6	97.8	97	97	97.4	96.2
M3DM2026.03		95.2	97.2	97.3	89.1	93.2	84.3	97	95.6	96.8	96.6	94.2
CMDR–IAD2026.03		95.2	96.3	83.1	96.9	97.5	85.1	90.7	98.1	96.3	96	93.5
CMDIAD2026.03		95.1	97.2	97.3	89.1	93.2	84.3	97	95.6	96.8	96.6	94.2
3D-ST2026.03		95	48.3	98.6	92.1	90.5	63.2	94.5	98.8	97.6	54.2	83.3
PatchCore2026.03		90.1	94.9	92.8	87.7	89.2	56.3	90.4	93.2	90.8	90.6	87.6
BTF2026.03		90.1	94.9	92.8	87.7	89.2	56.3	90.4	93.2	90.8	90.6	87.6
BTM2026.03		90.1	95.8	94.2	90.5	95.1	61.5	90.6	93.8	92.7	91.6	89.6
AST2026.03		83.4	87.4	76.4	44.7	87.2	56.9	84.5	51.8	86.3	58.8	71.7
Voxel GAN2026.03		66.4	62	76.6	74	78.3	33.2	58.2	79	63.3	48.3	63.9
Voxel VM2026.03		51	33.1	41.3	71.5	68	27.9	30	50.7	61.1	36.6	47.1
Voxel AE2026.03		46.7	75	80.8	55	76.5	47.3	72.1	91.8	1.9	17	56.4
Depth AE2026.03		43.2	15.8	80.8	49.1	84.1	40.6	26.2	21.6	71.6	47.8	48.1
Depth GAN2026.03		42.1	42.2	77.8	69.6	49.4	25.2	28.5	36.2	40.2	63.1	47.4
Depth VM2026.03		38.8	32.1	19.4	57	40.8	28.2	24.4	34.9	26.8	33.1	33.5