Quantifying and Detecting Collective Motion in Crowd Scenes

People in crowd scenes always exhibit consistent behaviors and form collective motions. The analysis of collective motion has motivated a surge of interest in computer vision. Nevertheless, the effort is hampered by the complex nature of collective motions. Considering the fact that collective motions are formed by individuals, this paper proposes a new framework for both quantifying and detecting collective motion by investigating the spatio-temporal behavior of individuals. The main contributions of this work are threefold: 1) an intention-aware model is built to fully capture the intrinsic dynamics of individuals; 2) a structure-based collectiveness measurement is developed to accurately quantify the collective properties of crowds; 3) a multi-stage clustering strategy is formulated to detect both the local and global behavior consistency in crowd scenes. Experiments on real world data sets show that our method is able to handle crowds with various structures and time-varying dynamics. Especially, the proposed method shows nearly 10% improvement over the competitors in terms of NMI, Purity and RI. Its applicability is illustrated in the context of anomaly detection and semantic scene segmentation.

I. Introduction

Collective motion, which is the primary component that makes up a crowd, is one of the most attractive phenomena in both nature and human society. Individuals in a collective motion tend to share consistent property, which is fundamentally important for analyzing the underlying pattern of crowd behavior. Since collective motion provides a mid-level representation of crowds, it has drawn increasing attentions in the field of computer vision, and involves a wide range of applications, such as crowd tracking [1]–[3], crowd counting [4], [5] and action recognition [6]–[9]. However, due to the complex spatial distribution and time-varying dynamics in crowd scenes, both the quantification and detection of collective motion are still difficult tasks.