PANDA-OE

Reported on 5 benchmarks across 1 task · 1 paper · 3 SOTA

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

Methodology5 results

Anomaly DetectiononOne-class CIFAR-100
AUROC· uses extra data· 2020-10-12
97.3
best: 98.4 (GeneralAD)
SOTA
PANDA: Adapting Pretrained Features for Anomaly Detection and Segmentation arXiv:2010.05903
Anomaly DetectiononOne-class CIFAR-10
AUROC· uses extra data· 2020-10-12
98.9
best: 99.6 (CLIP (OE))
SOTA
PANDA: Adapting Pretrained Features for Anomaly Detection and Segmentation arXiv:2010.05903
Anomaly DetectiononDIOR
ROC AUC· 2020-10-12
95.9
SOTA
PANDA: Adapting Pretrained Features for Anomaly Detection and Segmentation arXiv:2010.05903
Anomaly DetectiononFashion-MNIST
ROC AUC· 2020-10-12
91.8
best: 98.6 (GAN based Anomaly Detection in Imbalance Problems)
PANDA: Adapting Pretrained Features for Anomaly Detection and Segmentation arXiv:2010.05903
Anomaly DetectiononCats-and-Dogs
ROC AUC· 2020-10-12
94.5
best: 97.3 (PANDA)
PANDA: Adapting Pretrained Features for Anomaly Detection and Segmentation arXiv:2010.05903