Context-aware Video Anomaly Detection in Long-Term Datasets

arXiv:2404.07887v1 Announce Type: new
Abstract: Video anomaly detection research is generally evaluated on short, isolated benchmark videos only a few minutes long. However, in real-world environments, security cameras observe the same scene for months or years at a time, and the notion of anomalous behavior critically depends on context, such as the time of day, day of week, or schedule of events. Here, we propose a context-aware video anomaly detection algorithm, Trinity, specifically targeted to these scenarios. Trinity is especially …

abstract anomaly anomaly detection arxiv behavior benchmark cameras context cs.cv datasets detection environments however long-term notion observe research security security cameras type video videos world