ComplexVAD: Detecting Interaction Anomalies in Video
This work addresses a gap in video anomaly detection research for scenarios involving complex object interactions, though it is incremental as it builds on existing methods with a new dataset and modeling approach.
The authors tackled the problem of detecting complex anomalies in videos, which involve interactions between objects, by introducing a new dataset called ComplexVAD and proposing a novel method that models object interactions using scene graphs with spatio-temporal attributes, resulting in their method outperforming existing state-of-the-art approaches on this dataset.
Existing video anomaly detection datasets are inadequate for representing complex anomalies that occur due to the interactions between objects. The absence of complex anomalies in previous video anomaly detection datasets affects research by shifting the focus onto simple anomalies. To address this problem, we introduce a new large-scale dataset: ComplexVAD. In addition, we propose a novel method to detect complex anomalies via modeling the interactions between objects using a scene graph with spatio-temporal attributes. With our proposed method and two other state-of-the-art video anomaly detection methods, we obtain baseline scores on ComplexVAD and demonstrate that our new method outperforms existing works.