VCP-CLIP

Reported on 5 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.

Methodology5 results

Anomaly DetectiononMVTec AD
Segmentation AP· 2024-07-17
49.4
best: 87.6 (WeakREST-Block)
VCP-CLIP: A visual context prompting model for zero-shot anomaly segmentation arXiv:2407.12276
Anomaly DetectiononMVTec AD
Segmentation AUPRO· 2024-07-17
87.3
best: 98.4 (WeakREST-Block)
VCP-CLIP: A visual context prompting model for zero-shot anomaly segmentation arXiv:2407.12276
Anomaly DetectiononMVTec AD
Segmentation AUROC· 2024-07-17
92
best: 99.7 (WeakREST-Block)
VCP-CLIP: A visual context prompting model for zero-shot anomaly segmentation arXiv:2407.12276
Anomaly DetectiononVisA
Segmentation AUPRO· uses extra data· 2024-07-17
90.7
best: 96 (DiffusionAD)
VCP-CLIP: A visual context prompting model for zero-shot anomaly segmentation arXiv:2407.12276
Anomaly DetectiononVisA
Segmentation AUROC· uses extra data· 2024-07-17
95.7
best: 99.1 (Dinomaly ViT-L (model-unified multi-class))
VCP-CLIP: A visual context prompting model for zero-shot anomaly segmentation arXiv:2407.12276