MemSeg

Reported on 3 benchmarks across 1 task · 1 paper · 2 SOTA

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

Methodology3 results

Anomaly DetectiononMVTec AD
FPS· uses extra data· 2022-05-02
31.3
best: 1016 (CPR-faster(TensorRT))
SOTA
MemSeg: A semi-supervised method for image surface defect detection using differences and commonalities arXiv:2205.00908
Anomaly DetectiononMVTec AD
Segmentation AUROC· uses extra data· 2022-05-02
98.84
best: 99.7 (WeakREST-Block)
SOTA
MemSeg: A semi-supervised method for image surface defect detection using differences and commonalities arXiv:2205.00908
Anomaly DetectiononMVTec AD
Detection AUROC· uses extra data· 2022-05-02
99.56
best: 99.9 (GLASS)
MemSeg: A semi-supervised method for image surface defect detection using differences and commonalities arXiv:2205.00908