Human Kinematics-inspired Skeleton-based Video Anomaly Detection

Jian Xiao,Tianyuan Liu,Genlin Ji

DOI: https://doi.org/10.48550/arXiv.2309.15662

2023-09-27

Abstract:Previous approaches to detecting human anomalies in videos have typically relied on implicit modeling by directly applying the model to video or skeleton data, potentially resulting in inaccurate modeling of motion information. In this paper, we conduct an exploratory study and introduce a new idea called HKVAD (Human Kinematic-inspired Video Anomaly Detection) for video anomaly detection, which involves the explicit use of human kinematic features to detect anomalies. To validate the effectiveness and potential of this perspective, we propose a pilot method that leverages the kinematic features of the skeleton pose, with a specific focus on the walking stride, skeleton displacement at feet level, and neck level. Following this, the method employs a normalizing flow model to estimate density and detect anomalies based on the estimated density. Based on the number of kinematic features used, we have devised three straightforward variant methods and conducted experiments on two highly challenging public datasets, ShanghaiTech and UBnormal. Our method achieves good results with minimal computational resources, validating its effectiveness and potential.

Computer Vision and Pattern Recognition

What problem does this paper attempt to address?

The problem that this paper attempts to solve is the inaccuracy of existing methods due to implicit modeling of motion information when detecting human abnormal behaviors in videos. Specifically, existing methods based on video or skeleton data may not accurately capture the complexity of human motion, thus affecting the effect of anomaly detection. For example, these methods may wrongly recognize the behavior of maintaining an upright posture while skateboarding (regarded as an abnormal event in the ShanghaiTech dataset) as normal walking.

To solve this problem, the paper proposes a new method named HKV AD (Human Kinematic - inspired Video Anomaly Detection), which detects anomalies by explicitly using human kinematic features. Specifically, HKV AD utilizes kinematic features such as stride, foot displacement, and neck displacement in skeleton postures, and adopts a normalizing flow model to estimate density, and then detects anomalies according to the estimated density.

The paper verifies the effectiveness and potential of this method by conducting experiments on two challenging public datasets (ShanghaiTech and UBnormal). The experimental results show that even with limited computing resources, HKV AD can achieve good results.