Contracting skeletal kinematics for human-related video anomaly detection

• Published in: Pattern recognition Vol. 156; p. 110817
• Main Authors: Flaborea, Alessandro, D’Amely di Melendugno, Guido Maria, D’Arrigo, Stefano, Sterpa, Marco Aurelio, Sampieri, Alessio, Galasso, Fabio
• Format: Journal Article
• Language: English
• Published: Elsevier Ltd 01-12-2024
• Subjects: Anomaly detection; Graph convolutional networks; Hyperbolic geometry; Kinematic skeleton; Open set recognition; Hyperbolic geometry; Kinematic skeleton; Graph convolutional networks; Open set recognition; Anomaly detection
• ISSN: 0031-3203
• DOI: 10.1016/j.patcog.2024.110817

Detecting the anomaly of human behavior is paramount to timely recognizing endangering situations, such as street fights or elderly falls. However, anomaly detection is complex since anomalous events are rare and because it is an open set recognition task, i.e., what is anomalous at inference has not been observed at training. We propose COSKAD, a novel model that encodes skeletal human motion by a graph convolutional network and learns to COntract SKeletal kinematic embeddings onto a latent hypersphere of minimum volume for Video Anomaly Detection. We propose three latent spaces: the commonly-adopted Euclidean and the novel spherical and hyperbolic. All variants outperform the state-of-the-art on the most recent UBnormal dataset, for which we contribute a human-related version with annotated skeletons. COSKAD sets a new state-of-the-art on the human-related versions of ShanghaiTech Campus and CUHK Avenue, with performance comparable to video-based methods. Source code and dataset will be released upon acceptance.
•We propose COSKAD: COntracting SKeletal kinematic embedding for HR video Anomaly Detection.•COSKAD explores three metric spaces where the latent distribution is altered.•State-of-the-art performance on three benchmarks: HR-UBnormal, HR-Avenue, and HR-STC.•We introduce a new skeleton-based Human-Related benchmark, HR-UBnormal.•The paper presents an in-depth analysis of the components and manifolds of COSKAD.