Anomaly Detection On One Class Cifar 10

评估指标

AUROC

评测结果

各个模型在此基准测试上的表现结果

模型名称	AUROC	Paper Title	Repository
GAN based Anomaly Detection in Imbalance Problems	90.6	GAN-based Anomaly Detection in Imbalance Problems	-
FCDD	92	Explainable Deep One-Class Classification	-
IGD (pre-trained SSL)	91.25	Deep One-Class Classification via Interpolated Gaussian Descriptor	-
GeneralAD	99.3	GeneralAD: Anomaly Detection Across Domains by Attending to Distorted Features	-
GOAD	88.2	Classification-Based Anomaly Detection for General Data	-
DINO-FT	98.4	Anomaly Detection Requires Better Representations	-
SSOOD	90.1	Using Self-Supervised Learning Can Improve Model Robustness and Uncertainty	-
CSI	94.3	CSI: Novelty Detection via Contrastive Learning on Distributionally Shifted Instances	-
ARNET	86.6	Attribute Restoration Framework for Anomaly Detection	-
PANDA	96.2	PANDA: Adapting Pretrained Features for Anomaly Detection and Segmentation	-
DN2	92.5	Deep Nearest Neighbor Anomaly Detection	-
IGD (scratch)	74.33	Deep One-Class Classification via Interpolated Gaussian Descriptor	-
DisAug CLR	92.5	Learning and Evaluating Representations for Deep One-class Classification	-
SSD	90.0	SSD: A Unified Framework for Self-Supervised Outlier Detection	-
Self-Supervised One-class SVM, RBF kernel	64.7	PANDA: Adapting Pretrained Features for Anomaly Detection and Segmentation	-
OLED	67.1	OLED: One-Class Learned Encoder-Decoder Network with Adversarial Context Masking for Novelty Detection	-
Transformaly	98.3	Transformaly -- Two (Feature Spaces) Are Better Than One	-
Mean-Shifted Contrastive Loss	98.6	Mean-Shifted Contrastive Loss for Anomaly Detection	-
UTAD	88.4	Unsupervised Two-Stage Anomaly Detection	-
CLIP (OE)	99.6	Exposing Outlier Exposure: What Can Be Learned From Few, One, and Zero Outlier Images	-

0 of 36 row(s) selected.