Dynamic Distinction Learning: Adaptive Pseudo Anomalies for Video Anomaly Detection

arXiv:2404.04986v1 Announce Type: new
Abstract: We introduce Dynamic Distinction Learning (DDL) for Video Anomaly Detection, a novel video anomaly detection methodology that combines pseudo-anomalies, dynamic anomaly weighting, and a distinction loss function to improve detection accuracy. By training on pseudo-anomalies, our approach adapts to the variability of normal and anomalous behaviors without fixed anomaly thresholds. Our model showcases superior performance on the Ped2, Avenue and ShanghaiTech datasets, where individual models are tailored for each scene. These achievements highlight DDL's effectiveness …

abstract accuracy anomaly anomaly detection arxiv cs.ai cs.cv detection dynamic function loss methodology normal novel training type video