TasksSotADatasetsPapersMethodsSubmitAbout
Papers With Code 2

A community resource for machine learning research: papers, code, benchmarks, and state-of-the-art results.

Explore

Notable BenchmarksAll SotADatasetsPapersMethods

Community

Submit ResultsAbout

Data sourced from the PWC Archive (CC-BY-SA 4.0). Built by the community, for the community.

Models/MuSc (zero-shot)

MuSc (zero-shot)

Reported on 16 benchmarks across 1 task · 1 paper

Note: results are matched by exact model name. Different papers may use the same name for different model variants.

Methodology16 results

  • Anomaly DetectiononBTAD
    Detection AUROC· 2024-01-30
    96.16
    best: 98.3 (CPR)
    MuSc: Zero-Shot Industrial Anomaly Classification and Segmentation with Mutual Scoring of the Unlabeled ImagesarXiv:2401.16753
  • Anomaly DetectiononBTAD
    Segmentation AP· 2024-01-30
    57.15
    best: 84.6 (WeakREST-Block)
    MuSc: Zero-Shot Industrial Anomaly Classification and Segmentation with Mutual Scoring of the Unlabeled ImagesarXiv:2401.16753
  • Anomaly DetectiononBTAD
    Segmentation AUPRO· 2024-01-30
    83.43
    best: 97.5 (ReConPatch WRN-50)
    MuSc: Zero-Shot Industrial Anomaly Classification and Segmentation with Mutual Scoring of the Unlabeled ImagesarXiv:2401.16753
  • Anomaly DetectiononBTAD
    Segmentation AUROC· 2024-01-30
    97.35
    best: 99.3 (WeakREST-Block)
    MuSc: Zero-Shot Industrial Anomaly Classification and Segmentation with Mutual Scoring of the Unlabeled ImagesarXiv:2401.16753
  • Anomaly DetectiononMVTec AD
    Detection AUROC· uses extra data· 2024-01-30
    97.8
    best: 99.9 (GLASS)
    MuSc: Zero-Shot Industrial Anomaly Classification and Segmentation with Mutual Scoring of the Unlabeled ImagesarXiv:2401.16753
  • Anomaly DetectiononMVTec AD
    Segmentation AP· uses extra data· 2024-01-30
    62.7
    best: 87.6 (WeakREST-Block)
    MuSc: Zero-Shot Industrial Anomaly Classification and Segmentation with Mutual Scoring of the Unlabeled ImagesarXiv:2401.16753
  • Anomaly DetectiononMVTec AD
    Segmentation AUPRO· uses extra data· 2024-01-30
    93.8
    best: 98.4 (WeakREST-Block)
    MuSc: Zero-Shot Industrial Anomaly Classification and Segmentation with Mutual Scoring of the Unlabeled ImagesarXiv:2401.16753
  • Anomaly DetectiononMVTec AD
    Segmentation AUROC· uses extra data· 2024-01-30
    97.3
    best: 99.7 (WeakREST-Block)
    MuSc: Zero-Shot Industrial Anomaly Classification and Segmentation with Mutual Scoring of the Unlabeled ImagesarXiv:2401.16753
  • Anomaly DetectiononVisA
    Detection AUROC· 2024-01-30
    92.8
    best: 99.8 (UniNet)
    MuSc: Zero-Shot Industrial Anomaly Classification and Segmentation with Mutual Scoring of the Unlabeled ImagesarXiv:2401.16753
  • Anomaly DetectiononVisA
    Segmentation AUPRO· 2024-01-30
    92.7
    best: 96 (DiffusionAD)
    MuSc: Zero-Shot Industrial Anomaly Classification and Segmentation with Mutual Scoring of the Unlabeled ImagesarXiv:2401.16753
  • Anomaly DetectiononVisA
    Segmentation AUPRO (until 30% FPR)· 2024-01-30
    92.7
    best: 96.1 (AnomalyDINO-S (full-shot))
    MuSc: Zero-Shot Industrial Anomaly Classification and Segmentation with Mutual Scoring of the Unlabeled ImagesarXiv:2401.16753
  • Anomaly DetectiononVisA
    Segmentation AUROC· 2024-01-30
    98.8
    best: 99.1 (Dinomaly ViT-L (model-unified multi-class))
    MuSc: Zero-Shot Industrial Anomaly Classification and Segmentation with Mutual Scoring of the Unlabeled ImagesarXiv:2401.16753
  • Anomaly DetectiononMVTec LOCO AD
    Avg. Detection AUROC· 2024-01-30
    75.9
    best: 95.3 (CSAD)
    MuSc: Zero-Shot Industrial Anomaly Classification and Segmentation with Mutual Scoring of the Unlabeled ImagesarXiv:2401.16753
  • Anomaly DetectiononMVTec LOCO AD
    Detection AUROC (only logical)· 2024-01-30
    67.47
    best: 98.1 (PSAD)
    MuSc: Zero-Shot Industrial Anomaly Classification and Segmentation with Mutual Scoring of the Unlabeled ImagesarXiv:2401.16753
  • Anomaly DetectiononMVTec LOCO AD
    Detection AUROC (only structural)· 2024-01-30
    84.3
    best: 95.9 (PUAD-M)
    MuSc: Zero-Shot Industrial Anomaly Classification and Segmentation with Mutual Scoring of the Unlabeled ImagesarXiv:2401.16753
  • Anomaly DetectiononMVTec LOCO AD
    Segmentation AU-sPRO (until FPR 5%)· 2024-01-30
    63.04
    best: 83.2 (SAM-LAD)
    MuSc: Zero-Shot Industrial Anomaly Classification and Segmentation with Mutual Scoring of the Unlabeled ImagesarXiv:2401.16753