Understanding crowd flow patterns using active-Langevin model

作者：

Highlights：

• The human crowd being considered as active colloid particles similar to the fluid.

• Active Langevin model to describe the crowd behavior.

• Segments linear and non-linear ordered and unordered crowd flows.

• The method outperforms existing popular methods in terms of accuracy and NMI scores.

摘要

•The human crowd being considered as active colloid particles similar to the fluid.•Active Langevin model to describe the crowd behavior.•Segments linear and non-linear ordered and unordered crowd flows.•The method outperforms existing popular methods in terms of accuracy and NMI scores.

论文关键词：Visual surveillance,Active Langevin equation,Crowd analysis,Human flow segmentation,Dense crowd

论文评审过程：Received 28 September 2020, Revised 18 February 2021, Accepted 11 May 2021, Available online 23 May 2021, Version of Record 4 June 2021.

论文官网地址：https://doi.org/10.1016/j.patcog.2021.108037