Towards the cognitive and psychological perspectives of crowd behaviour: a vision-based analysis

Abstract

Smart and proactive surveillance of crowds has turned up into greater importance in recent years due to the increase in urban population and immutable crowd disasters. Although the behaviours that emerged in a crowd are often unpredictable, researches in computer vision try to figure out this emergent behaviour based on the psychological and cognitive aspects of the crowd. This review is intended to analyse the insights shared from these aspects for analysing crowd behaviour. We also try to uncover the crowd psychological theories present in the literature that largely helps in determining the non-adaptive crowd behaviours. Further, this paper discusses an outlook of state-of-the-art psychological, cognitive computing and cognitive psychological approaches of crowd behaviour analysis. This paper also provides a discussion on the benchmark datasets available for vision-based crowd analysis. The comprehensive discussion and insights based on crowd psychology and cognition will provide a deep understanding of the fundamental prospects of smart crowd behaviour analysis.