Anomaly Detection on Inline-Food-Inspection-THz one-class setting

0.986Total AUROCBeer-Lambert guided representation learning

Evaluation Results

Method
Beer-Lambert guided representation learning2026.06	0.986	0.951	0.93	1	0.986	0.993	0.972	1	0.986	0.968
EfficientAD2026.06	0.983	0.94	0.907	0.993	0.959	1	0.986	1	0.986	0.959
PatchCore2026.06	0.977	0.993	0.959	0.961	0.959	0.956	0.959	1	0.986	0.966
EfficientAD2026.06	0.975	—	—	—	—	—	—	—	—	0.925
Beer-Lambert guided representation learning2026.06	0.935	—	—	—	—	—	—	—	—	0.917
PaDiM2026.06	0.93	0.949	0.972	0.851	0.932	0.924	0.959	0.998	0.972	0.959
Reverse Distillation2026.06	0.916	0.993	0.959	0.991	0.972	0.679	0.843	1	0.986	0.94
Beer-Lambert guided representation learning2026.06	0.884	—	—	—	—	—	—	—	—	0.879
EfficientAD2026.06	0.865	—	—	—	—	—	—	—	—	0.84
Reverse Distillation2026.06	0.829	—	—	—	—	—	—	—	—	0.753
PatchCore2026.06	0.812	—	—	—	—	—	—	—	—	0.759
UniNet2026.06	0.744	0.898	0.907	0.694	0.864	0.697	0.843	0.687	0.843	0.864
PatchCore2026.06	0.743	—	—	—	—	—	—	—	—	0.725
PaDiM2026.06	0.725	—	—	—	—	—	—	—	—	0.757
Reverse Distillation2026.06	0.718	—	—	—	—	—	—	—	—	0.745
SupersimpleNet2026.06	0.67	—	—	—	—	—	—	—	—	0.715
PaDiM2026.06	0.622	—	—	—	—	—	—	—	—	0.712
SupersimpleNet2026.06	0.606	—	—	—	—	—	—	—	—	0.681
UniNet2026.06	0.441	—	—	—	—	—	—	—	—	0.654
UniNet2026.06	0.422	—	—	—	—	—	—	—	—	0.655
SupersimpleNet2026.06	0.326	0.641	0.843	0.047	0.843	0.567	0.843	0.05	0.843	0.843