uTRAND: Unsupervised Anomaly Detection in Traffic Trajectories

Giacomo D'amicantonio, Egor Bondarau, Peter H.N. De With; Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) Workshops, 2024, pp. 7638-7645

Deep learning-based approaches have achieved signif- icant improvements on public video anomaly datasets but often do not perform well in real-world applications. This paper addresses two issues: the lack of labeled data and the difficulty of explaining the predictions of a neural net- work. To this end we present a framework called uTRAND that shifts the problem of anomalous trajectory prediction from the pixel space to a semantic-topological domain. The framework detects and tracks all types of traffic agents in bird's-eye-view videos of traffic cameras mounted at an in- tersection. By conceptualizing the intersection as a patch- based graph it is shown that the framework learns and models the normal behaviour of traffic agents without costly manual labeling. Furthermore uTRAND allows to formu- late simple rules to classify anomalous trajectories in a way suited for human interpretation. We show that uTRAND outperforms other state-of-the-art approaches on a dataset of anomalous trajectories collected in a real-world setting while producing explainable detection results.