Learning and Evaluating Representations for Deep One-class Classification

Benchmark Model Rank Results
anomaly-detection-on-one-class-cifar-100DisAug CLR#7AUROC: 86.5
anomaly-detection-on-one-class-cifar-100Rotation Prediction#8AUROC: 84.1
anomaly-detection-on-one-class-cifar-10DisAug CLR#12AUROC: 92.5
anomaly-detection-on-mvtec-adDisAug CLR#103Detection AUROC: 86.5Segmentation AUROC: 90.4
anomaly-detection-on-mvtec-adRotNet (MLP Head)#104Detection AUROC: 86.3Segmentation AUROC: 93