Anomaly Detection on MVTecAD (test)
100Bottle ScoreSOMAD
Evaluation results
| Method | Links | ||||||||||||||||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| SOMAD2021.07 | 100 | — | — | — | — | — | — | — | — | 100 | 93.8 | 99.7 | 100 | 98.6 | 99.2 | 100 | 98.7 | 93.9 | 98.8 | 94.5 | 98.6 | 95.5 | 97.7 | — | 97.9 | — | |
| [20]Learning Mode=Unsupervised2021.08 | 100 | — | — | — | — | — | — | — | — | 98.7 | 92.3 | 93.1 | 100 | 83.4 | 97.6 | 95.5 | 97.4 | 92.9 | 94 | 95.9 | 95.8 | 81.2 | 97.9 | 94.4 | — | — | |
| CS-Flowevaluation_level=image-level, training_samples=16 shots2021.10 | 100 | — | — | — | — | — | — | — | — | 97.9 | 83.1 | 99.1 | 100 | 90.9 | 99.5 | 100 | 99.9 | 93.3 | 94.4 | 98 | 85.6 | 65.2 | 95.3 | 91 | — | 98.5 | |
| InTraPatch Size (K)=16, Patch Window (L)=72021.04 | 100 | — | — | — | — | — | — | — | 95 | 95.7 | 86.5 | 96.9 | 100 | 90.2 | 97.5 | 98.8 | 98.2 | 100 | 70.3 | 95.8 | 100 | 95.7 | 99.4 | 93 | — | 98.9 | |
| PMADMethod Category=Unified, Evaluation Setting=one-for-one2024.03 | 100 | — | — | — | — | — | — | — | — | 100 | 89.8 | 99.2 | 100 | 94.3 | 98 | 99.7 | 100 | 97.7 | 98 | 99.8 | 91.4 | 73.9 | 99.5 | 96.1 | — | — | |
| UniADMethod Category=Unified, Evaluation Setting=one-for-one2024.03 | 100 | — | — | — | — | — | — | — | — | 99.9 | 85.3 | 99 | 100 | 88.3 | 97.9 | 99.9 | 99 | 98.5 | 97.6 | 100 | 95 | 91.9 | 96.7 | 96.6 | — | — | |
| FastFlowMethod Category=NF Based, Evaluation Setting=one-for-one2024.03 | 100 | — | — | — | — | — | — | — | — | 100 | 100 | 100 | 100 | 99.4 | 100 | 100 | 100 | 99.7 | 100 | 99.8 | 94.4 | 97.8 | 99.5 | 99.4 | — | — | |
| CFLOWMethod Category=NF Based, Evaluation Setting=one-for-one2024.03 | 100 | — | — | — | — | — | — | — | — | 96.8 | 99.3 | 91.9 | 97.7 | 99.9 | 99.6 | 100 | 98.7 | 97.6 | 100 | 99.1 | 95.2 | 99.7 | 98.5 | 98.3 | — | — | |
| [31]Learning Mode=Unsupervised2021.08 | 99.9 | — | — | — | — | — | — | — | — | 83.3 | 88.4 | 88.5 | 100 | 83.8 | 93 | 84.2 | 98.7 | 99.6 | 81.9 | 90.9 | 100 | 84.5 | 98.1 | 91.7 | — | — | |
| PaDimLearning Mode=Unsupervised2021.08 | 99.9 | — | — | — | — | — | — | — | — | 92 | 91.3 | 98.7 | 100 | 93.3 | 99.2 | 99.8 | 98.1 | 96.7 | 92.7 | 97.4 | 95.1 | 85.8 | 90.3 | 95.5 | — | — | |
| RIAD2021.04 | 99.9 | — | — | — | — | — | — | — | 91.7 | 83.3 | 88.4 | 88.5 | 100 | 83.8 | 93 | 84.2 | 93.4 | 99.6 | 81.9 | 90.9 | 100 | 84.5 | 98.1 | 89.9 | — | 95.1 | |
| PaDiMMethod Category=Baseline, Evaluation Setting=one-for-one2024.03 | 99.9 | — | — | — | — | — | — | — | — | 92 | 91.3 | 98.7 | 100 | 93.3 | 99.2 | 99.8 | 98.1 | 96.7 | 92.7 | 97.4 | 96.1 | 85.8 | 90.3 | 95.5 | — | — | |
| PaDiM2021.07 | 99.8 | — | — | — | — | — | — | — | — | 93.3 | 91.5 | 99.2 | 100 | 94.4 | 98.8 | 99.9 | 97.4 | 95.7 | 92.2 | 97.8 | 97.2 | 84.4 | 90.9 | — | 95.5 | — | |
| CS-Flowevaluation_level=image-level, training_samples=full set2021.10 | 99.8 | — | — | — | — | — | — | — | — | 99.6 | 97.1 | 99.1 | 100 | 98.6 | 100 | 100 | 100 | 99 | 99.1 | 99.3 | 91.9 | 97.6 | 99.7 | 98.2 | — | 99.8 | |
| 1-NNevaluation_level=image-level, training_set=full set2021.10 | 99.6 | — | — | — | — | — | — | — | — | 98.2 | 89.4 | 91.1 | 100 | 79.9 | 95.8 | 98.5 | 100 | 81.8 | 95.1 | 95.4 | 94.7 | 91.4 | 97.1 | 93.2 | — | 96.1 | |
| MKDMethod Category=Baseline, Evaluation Setting=one-for-one2024.03 | 99.4 | — | — | — | — | — | — | — | — | 98.4 | 80.5 | 73.6 | 95.1 | 82.7 | 94.3 | 79.3 | 91.6 | 78 | 89.2 | 85.6 | 92.2 | 83.3 | 93.2 | 87.8 | — | — | |
| DRÆMLearning Mode=Unsupervised, epochs=700, learning rate=10-4, anomaly source=Describable Textures Dataset [9]2021.08 | 99.2 | — | — | — | — | — | — | — | — | 100 | 98.5 | 98.7 | 100 | 98.9 | 99.1 | 97 | 99.6 | 99.9 | 91.8 | 93.1 | 100 | 93.9 | 100 | 98 | — | — | |
| DRAEMMethod Category=Baseline, Evaluation Setting=one-for-one2024.03 | 99.2 | — | — | — | — | — | — | — | — | 100 | 98.5 | 98.7 | 100 | 98.9 | 99.1 | 97 | 99.6 | 99.9 | 91.8 | 93.1 | 100 | 93.9 | 100 | 98 | — | — | |
| Uninformed StudentsLearning Mode=Unsupervised2021.08 | 99 | — | — | — | — | — | — | — | — | 93.1 | 86.1 | 82 | 88.2 | 87.9 | 97.7 | 91.6 | 99.1 | 81 | 86.2 | 81.8 | 95.3 | 54.9 | 91.9 | 87.7 | — | — | |
| Mahal.evaluation_level=image-level, training_set=full set2021.10 | 99 | — | — | — | — | — | — | — | — | 100 | 96.3 | 97.4 | 100 | 91.4 | 99.3 | 99.6 | 100 | 93.7 | 99.1 | 98.2 | 94.1 | 94.5 | 98.8 | 96.9 | — | 98.5 | |
| DifferNetevaluation_level=image-level, training_set=full set2021.10 | 99 | — | — | — | — | — | — | — | — | 99.3 | 86.9 | 96.1 | 97.1 | 88.8 | 99.8 | 92.9 | 99.4 | 84 | 95.9 | 91.1 | 98.6 | 96.3 | 95.1 | 94.7 | — | 94.6 | |
| Patch SVDD2021.10 | 99 | — | — | — | — | — | — | — | — | 92 | 77 | 94 | 91 | 86 | 92 | 93 | 98 | 95 | 90 | 92 | 100 | 81 | 98 | 92 | — | — | |
| PatchSVDD2021.07 | 98.6 | — | — | — | — | — | — | — | — | 92 | 76.7 | 94 | 90.9 | 86.1 | 96.5 | 92.9 | 97.8 | 94.6 | 90.3 | 91.5 | 100 | 81.3 | 97.9 | — | 92.1 | — | |
| [26]Learning Mode=Unsupervised2021.08 | 98.3 | — | — | — | — | — | — | — | — | 100 | 68.7 | 81.5 | 94.4 | 76.8 | 97.9 | 90.3 | 96.1 | 86.7 | 66.5 | 79.4 | 95 | 100 | 78.1 | 87.3 | — | — | |
| CutPaste2021.04 | 98.3 | — | — | — | — | — | — | — | 95.2 | 97.3 | 96.2 | 99.3 | 100 | 92.4 | 98.6 | 93.1 | 93.4 | 99.9 | 80.6 | 95.5 | 98.3 | 86.3 | 99.4 | 94.3 | — | 97 | |
| Cutpaste2021.10 | 98 | — | — | — | — | — | — | — | — | 97 | 96 | 99 | 100 | 92 | 99 | 93 | 93 | 99 | 81 | 96 | 98 | 86 | 99 | 95 | — | — | |
| AnoSeg2021.10 | 98 | — | — | — | — | — | — | — | — | 98 | 84 | 95 | 99 | 87 | 99 | 96 | 98 | 99 | 98 | 96 | 99 | 97 | 99 | 96 | — | — | |
| SPADE2021.07 | 97.2 | — | — | — | — | — | — | — | — | 88.1 | 89.7 | 71 | 95.4 | 80.1 | 95.8 | 92.8 | 96.5 | 47.3 | 84.4 | 90.3 | 88.9 | 66.7 | 96.6 | — | 85.4 | — | |
| CAVGA-RwLearning Setting=Weakly Supervised, Latent Variable Configuration=Flattened fully connected2019.11 | 96 | — | — | — | — | — | — | — | — | 92 | 93 | 88 | 84 | 97 | 89 | 82 | 86 | 81 | 97 | 89 | 99 | 79 | 96 | 90 | — | — | |
| ARNetevaluation_level=image-level, training_set=full set2021.10 | 94.1 | — | — | — | — | — | — | — | — | 85.5 | 68.1 | 66.7 | 86.2 | 78.6 | 92.3 | 70.6 | 73.5 | 88.3 | 83.2 | 84.3 | 100 | 100 | 87.6 | 84.8 | — | 82.2 | |
| CAVGA-DwLearning Setting=Weakly Supervised, Latent Variable Configuration=Convolutional2019.11 | 93 | — | — | — | — | — | — | — | — | 90 | 89 | 81 | 80 | 93 | 89 | 80 | 81 | 79 | 86 | 80 | 96 | 79 | 95 | 86 | — | — | |
| CAVGA-RuLearning Setting=Unsupervised, Latent Variable Configuration=Flattened fully connected2019.11 | 91 | — | — | — | — | — | — | — | — | 87 | 87 | 71 | 75 | 91 | 88 | 78 | 72 | 78 | 67 | 75 | 97 | 78 | 94 | 82 | — | — | |
| CAVGA2021.10 | 91 | — | — | — | — | — | — | — | — | 87 | 87 | 71 | 75 | 91 | 88 | 78 | 72 | 78 | 67 | 75 | 97 | 78 | 94 | 82 | — | — | |
| MOCCAboundary_type=soft2020.12 | 90 | — | — | — | — | — | — | — | — | 77 | 77 | 85 | 96 | 81 | 97 | 81 | 80 | 85 | 72 | 77 | 93 | 82 | 78 | — | — | — | |
| MOCCAboundary_type=hard2020.12 | 90 | — | — | — | — | — | — | — | — | 76 | 75 | 80 | 91 | 69 | 94 | 74 | 78 | 76 | 68 | 81 | 91 | 80 | 78 | — | — | — | |
| GANevaluation_level=image-level, training_set=full set2021.10 | 89.2 | — | — | — | — | — | — | — | — | 78.5 | 73.2 | 70 | 84.2 | 74.3 | 83.4 | 69.9 | 79.4 | 70.8 | 75.7 | 79.2 | 65.3 | 74.6 | 74.5 | 75.5 | — | 77.5 | |
| CAVGA-DuLearning Setting=Unsupervised, Latent Variable Configuration=Convolutional2019.11 | 89 | — | — | — | — | — | — | — | — | 84 | 83 | 67 | 71 | 88 | 85 | 73 | 70 | 75 | 63 | 73 | 91 | 77 | 87 | 78 | — | — | |
| AVID2019.11 | 88 | — | — | — | — | — | — | — | — | 86 | 85 | 63 | 58 | 86 | 83 | 70 | 66 | 59 | 64 | 58 | 73 | 66 | 84 | 73 | — | — | |
| AESSIM2019.11 | 88 | — | — | — | — | — | — | — | — | 54 | 61 | 54 | 46 | 60 | 83 | 67 | 52 | 69 | 61 | 52 | 74 | 51 | 80 | 63 | — | — | |
| AESSIM2020.12 | 88 | — | — | — | — | — | — | — | — | 54 | 61 | 54 | 46 | 60 | 83 | 67 | 52 | 69 | 61 | 52 | 74 | 51 | 80 | — | — | — | |
| AVID2020.12 | 88 | — | — | — | — | — | — | — | — | 86 | 85 | 63 | 58 | 86 | 83 | 70 | 66 | 59 | 64 | 58 | 73 | 66 | 84 | — | — | — | |
| gamma-VAE2019.11 | 86 | — | — | — | — | — | — | — | — | 74 | 86 | 78 | 71 | 80 | 89 | 67 | 81 | 83 | 56 | 70 | 89 | 71 | 67 | 77 | — | — | |
| LSA2019.11 | 86 | — | — | — | — | — | — | — | — | 80 | 71 | 67 | 70 | 85 | 75 | 74 | 70 | 54 | 61 | 50 | 89 | 75 | 88 | 73 | — | — | |
| VAE-grad2020.12 | 86 | — | — | — | — | — | — | — | — | 74 | 86 | 78 | 71 | 80 | 89 | 67 | 81 | 83 | 56 | 70 | 89 | 71 | 67 | — | — | — | |
| US2021.10 | 85 | — | — | — | — | — | — | — | — | 91 | 82 | 58 | 73 | 90 | 68 | 86 | 87 | 60 | 90 | 85 | 81 | 90 | 90 | 84 | — | — | |
| CBIGAN2020.12 | 84 | — | — | — | — | — | — | — | — | 75 | 58 | 67 | 87 | 76 | 88 | 60 | 84 | 99 | 73 | 74 | 97 | 67 | 55 | — | — | — | |
| DSEBMevaluation_level=image-level, training_set=full set2021.10 | 81.8 | — | — | — | — | — | — | — | — | 76.2 | 59.4 | 67.9 | 41.6 | 80.6 | 95.2 | 41.3 | 69 | 71.7 | 68.5 | 74.1 | 78.1 | 99.9 | 58.4 | 74.5 | — | 63.8 | |
| AEL22019.11 | 80 | — | — | — | — | — | — | — | — | 88 | 62 | 73 | 44 | 62 | 74 | 50 | 77 | 78 | 56 | 71 | 98 | 69 | 80 | 71 | — | — | |
| AE_L22021.10 | 80 | — | — | — | — | — | — | — | — | 88 | 62 | 73 | 44 | 62 | 74 | 50 | 77 | 78 | 56 | 71 | 98 | 69 | 80 | 71 | — | — | |
| AEL22020.12 | 80 | — | — | — | — | — | — | — | — | 88 | 62 | 73 | 44 | 62 | 74 | 50 | 77 | 78 | 56 | 71 | 98 | 69 | 80 | — | — | — | |
| [1]Learning Mode=Unsupervised2021.08 | 79.4 | — | — | — | — | — | — | — | — | 87.4 | 72.1 | 69.4 | 80.8 | 67.1 | 92 | 82.1 | 72 | 74.3 | 71.1 | 80.8 | 70 | 100 | 74.4 | 78.2 | — | — | |
| Geom.evaluation_level=image-level, training_set=full set2021.10 | 74.4 | — | — | — | — | — | — | — | — | 35.9 | 67 | 81.3 | 84.1 | 63 | 61.1 | 43.7 | 41.7 | 61.9 | 78.3 | 86.9 | 97.2 | 50 | 82 | 71.6 | — | 59.6 | |
| AnoGAN2019.11 | 69 | — | — | — | — | — | — | — | — | 50 | 58 | 50 | 52 | 62 | 68 | 49 | 51 | 51 | 53 | 67 | 57 | 35 | 59 | 55 | — | — | |
| AnoGAN2020.12 | 69 | — | — | — | — | — | — | — | — | 50 | 58 | 50 | 52 | 62 | 68 | 49 | 51 | 51 | 53 | 67 | 57 | 35 | 59 | — | — | — | |
| EGBAD2020.12 | 68 | — | — | — | — | — | — | — | — | 50 | 55 | 55 | 65 | 63 | 80 | 60 | 73 | 50 | 66 | 68 | 48 | 50 | 59 | — | — | — | |
| AnoSeg (+ Feature matching loss)Ablation=Base model + Feature matching loss2021.10 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | 84.2 | — | — | |
| AnoSeg (+ Input anomaly map)Ablation=Base model + Feature matching loss + Input anomaly map2021.10 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | 94.3 | — | — | |
| AnoSeg (Base model)Ablation=Base model2021.10 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | 81.2 | — | — | |
| BGADKnown Class=Carpet, Anomaly Type=Mean, Training Anomaly Samples=102022.07 | — | 99.9 | — | — | 99.5 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| BGADKnown Class=Metal nut, Anomaly Type=Mean, Training Anomaly Samples=102022.07 | — | 99.8 | — | — | 97.5 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| BGADLearning setting=Supervised AD2022.07 | — | 99.3 | — | — | 99.2 | — | — | 97.6 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| BGAD w/oLearning setting=Unsupervised AD, explicit boundary guidance=false2022.07 | — | 97.4 | — | — | 98.2 | — | — | 95.5 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| CFA*Learning setting=Unsupervised AD2022.07 | — | 98.9 | — | — | 98.2 | — | — | 93.1 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| CflowScenario=One-shot2024.03 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | 69 | — | |
| CflowScenario=Few-shot2024.03 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | 79.8 | — | |
| CflowScenario=Full-shot2024.03 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | 88.8 | — | |
| DeSTSegsetting=multi-class2025.04 | — | 89.2 | 95.5 | 91.6 | 93.1 | 54.3 | 50.9 | 64.8 | 77.1 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| DevNetKnown Class=Carpet, Anomaly Type=Mean, Training Anomaly Samples=102022.07 | — | 84.7 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| DevNetKnown Class=Metal nut, Anomaly Type=Mean, Training Anomaly Samples=102022.07 | — | 96.5 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| DIADsetting=multi-class, FPS=0.12025.04 | — | 97.2 | 99 | 96.5 | 96.8 | 52.6 | 55.5 | 90.7 | 84 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| DRAKnown Class=Carpet, Anomaly Type=Mean, Training Anomaly Samples=102022.07 | — | 93.5 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| DRAKnown Class=Metal nut, Anomaly Type=Mean, Training Anomaly Samples=102022.07 | — | 94.5 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| DRAEM*Learning setting=Unsupervised AD2022.07 | — | 97.8 | — | — | 96.9 | — | — | 94.7 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| Efficient-ADScenario=One-shot2024.03 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | 65.6 | — | |
| Efficient-ADScenario=Few-shot2024.03 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | 71.3 | — | |
| Efficient-ADScenario=Full-shot2024.03 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | 92.8 | — | |
| FLOSKnown Class=Carpet, Anomaly Type=Mean, Training Anomaly Samples=102022.07 | — | 76.1 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| FLOSKnown Class=Metal nut, Anomaly Type=Mean, Training Anomaly Samples=102022.07 | — | 92.2 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| HVQ-Transsetting=multi-class, FPS=5.62025.04 | — | 98 | 99.5 | 97.5 | 97.3 | 48.2 | 53.3 | 91.4 | 83.6 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| InvADsetting=multi-class, FPS=88.12025.04 | — | 99 | 99.6 | 98.5 | 97.5 | 46.5 | 52.3 | 92.7 | 83.7 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| MDMsetting=multi-class, FPS=1.92025.04 | — | 87.2 | 94.2 | 91 | 94.8 | 39.5 | 44.8 | 88 | 77.1 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| MLEPKnown Class=Carpet, Anomaly Type=Mean, Training Anomaly Samples=102022.07 | — | 75.1 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| MLEPKnown Class=Metal nut, Anomaly Type=Mean, Training Anomaly Samples=102022.07 | — | 87.8 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| MSFD*Learning setting=Unsupervised AD2022.07 | — | 96.4 | — | — | 97 | — | — | 91.5 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| NFADKnown Class=Carpet, Anomaly Type=Mean, Training Anomaly Samples=Baseline2022.07 | — | 99.8 | — | — | 99.4 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| NFADKnown Class=Metal nut, Anomaly Type=Mean, Training Anomaly Samples=Baseline2022.07 | — | 98.3 | — | — | 96.6 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| NFAD+Learning setting=Unsupervised AD2022.07 | — | 96.8 | — | — | 97.9 | — | — | 94.6 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| OmiADsetting=multi-class, FPS=39.42025.04 | — | 98.8 | 99.7 | 98.5 | 97.7 | 52.6 | 56.7 | 93.2 | 85.3 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| PaDiMevaluation_level=image-level, training_set=full set2021.10 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | 97.2 | — | 99 | |
| PaDiM*Learning setting=Unsupervised AD2022.07 | — | 97.5 | — | — | 97.6 | — | — | 93.2 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| PatchcoreScenario=One-shot2024.03 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | 76.9 | — | |
| PatchcoreScenario=Few-shot2024.03 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | 85.8 | — | |
| PatchcoreScenario=Full-shot2024.03 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | 97.3 | — | |
| PatchCore*Learning setting=Unsupervised AD2022.07 | — | 98.8 | — | — | 98.1 | — | — | 94 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| RD4ADsetting=multi-class, FPS=4.82025.04 | — | 94.6 | 96.5 | 95.2 | 96.1 | 48.6 | 53.8 | 91.1 | 82.3 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| Reverse-distillationScenario=One-shot2024.03 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | 48.4 | — | |
| Reverse-distillationScenario=Few-shot2024.03 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | 50.6 | — | |
| Reverse-distillationScenario=Full-shot2024.03 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | 79.2 | — | |
| SAOEKnown Class=Carpet, Anomaly Type=Mean, Training Anomaly Samples=102022.07 | — | 76.2 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| SAOEKnown Class=Metal nut, Anomaly Type=Mean, Training Anomaly Samples=102022.07 | — | 85.5 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — |