Visualization of Interactions in Crowd Simulation and Video Sequences

Abstract

Although crowd behavior has been investigated in several applications and a variety of purposes, just a few of the existing simulation methods take into account the phenomenon of interaction between persons. This work aims to use BioCrowds, endowing our agents with personalities and the ability to interact with each other, as well to design interactive visualizations which show relevant information about such simulations. Examples of visualization data is the occurrence of interactions as a function of personalities. Also, we extract such interactions between pedestrians from reallife video sequences, and visualize the output achieved with our visualization tool. The achieved results show that our agents are able to interact with each other as expected. Also, the designed visualizations were helpful to generate relevant information about the captured data, both from simulations and video sequences