Anomaly detection aims to separate anomalies from normal samples, and the
pretrained network is promising for anomaly detection. However, adapting the
pretrained features would be confronted with the risk of pattern collapse when
finetuning on one-class training data. In this paper, we propose an anomaly
detection framework called constrained adaptive projection with pretrained
features (CAP). Combined with pretrained features, a simple linear projection
head applied on a specific input and its k most similar pretrained normal
representations is designed for feature adaptation, and a reformed
self-attention is leveraged to mine the inner-relationship among one-class
semantic features. A loss function is proposed to avoid potential pattern
collapse. Concretely, it considers the similarity between a specific data and
its corresponding adaptive normal representation, and incorporates a constraint
term slightly aligning pretrained and adaptive spaces. Our method achieves
state-ofthe-art anomaly detection performance on semantic anomaly detection and
sensory anomaly detection benchmarks including 96.5% AUROC on CIFAR- 100
dataset, 97.0% AUROC on CIFAR-10 dataset and 89.9% AUROC on MvTec dataset.

本文提出一种基于预训练特征的异常检测框架 (CAP)，该框架通过线性投影头将输入数据和最相似的预训练正常数据匹配，并采用改进的自注意力机制挖掘单类语义特征之间的内在关系，以避免潜在的模式折叠问题，实现了语义异常检测和感知异常检测基准上的最优性能，如 CIFAR-100 数据集上的 96.5％ AUROC，CIFAR-10 数据集上的 97.0％ AUROC 和 MvTec 数据集上的 89.9％ AUROC。