Abnormal crowd behavior detection using social force model

TLDR

The paper proposes a novel method to detect and localize abnormal crowd behaviors in videos using a Social Force model. The method overlays a particle grid on the video, advects it with optical flow, estimates interaction forces via the Social Force model, maps these forces to produce Force Flow, models normal behavior from spatio‑temporal volumes, and classifies frames with a bag‑of‑words approach to localize anomalies. Experiments on the University of Minnesota escape panic dataset and a challenging web‑based crowd dataset show that the proposed method successfully captures crowd dynamics and outperforms pure optical flow approaches.

Abstract

In this paper we introduce a novel method to detect and localize abnormal behaviors in crowd videos using Social Force model. For this purpose, a grid of particles is placed over the image and it is advected with the space-time average of optical flow. By treating the moving particles as individuals, their interaction forces are estimated using social force model. The interaction force is then mapped into the image plane to obtain Force Flow for every pixel in every frame. Randomly selected spatio-temporal volumes of Force Flow are used to model the normal behavior of the crowd. We classify frames as normal and abnormal by using a bag of words approach. The regions of anomalies in the abnormal frames are localized using interaction forces. The experiments are conducted on a publicly available dataset from University of Minnesota for escape panic scenarios and a challenging dataset of crowd videos taken from the web. The experiments show that the proposed method captures the dynamics of the crowd behavior successfully. In addition, we have shown that the social force approach outperforms similar approaches based on pure optical flow.