Toward Multi-class Anomaly Detection: Exploring Class-aware Unified Model against Inter-class Interference

arXiv:2403.14213v1 Announce Type: new
Abstract: In the context of high usability in single-class anomaly detection models, recent academic research has become concerned about the more complex multi-class anomaly detection. Although several papers have designed unified models for this task, they often overlook the utility of class labels, a potent tool for mitigating inter-class interference. To address this issue, we introduce a Multi-class Implicit Neural representation Transformer for unified Anomaly Detection (MINT-AD), which leverages the fine-grained category information in the training …

abstract academic academic research anomaly anomaly detection arxiv become class context cs.cv detection interference labels papers research type unified model usability utility