A Hierarchical Bayesian Model for Crowd Emotions

Abstract

Estimation of emotions is an essential aspect in developing intelligent systems intended for crowded environments. However, emotion estimation in crowds remains a challenging problem due to the complexity in which human emotions are manifested and the capability of a system to perceive them in such conditions. This paper proposes a hierarchical Bayesian model to learn in unsupervised manner the behavior of individuals and of the crowd as a single entity, and explore the relation between behavior and emotions to infer emotional states. Information about the motion patterns of individuals are described using a self-organizing map, and a hierarchical Bayesian network builds probabilistic models to identify behaviors and infer the emotional state of individuals and the crowd. This model is trained and tested using data produced from simulated scenarios that resemble real-life environments. The conducted experiments tested the efficiency of our method to learn, detect and associate behaviors with emotional states yielding accuracy levels of 74% for individuals and 81% for the crowd, similar in performance with existing methods for pedestrian behavior detection but with novel concepts regarding the analysis of crowds.

Stress Propagation in Human-Robot Teams Based on Computational Logic Model

“…1. Development of a composite model is needed since the known models, e.g., [54], [62], [66], [70], do not satisfy the requirements of new realities such as teaming of humans and autonomous machines. In addition, wearable technologies based on a personalized collection of physiological and psychological data, e.g., [59], provide new possibilities for mission monitoring and control.…”

“…Modelling and simulation is a common way of understanding and predicting the behaviour and performance of groups of individuals [13]. Various approaches and methodologies for modelling the crowd behaviour, in particular, were considered in [41], [66], [67]. Integration of an agent-based model into an epidemic model provides useful insights [14].…”

“…Message-passing method overcomes these limitations. Stress (emotion) propagation similar to infection propagation can be simulated using Bayesian network [66], [70], epidemiological model using differential equations, e.g., SIS and SIR [47], message-passing algorithm [31], [32], as well as composite model [64] in which differential equations are used for the messages over time. Propagation using differential equations for the messages have been proposed in [64].…”

Emotion estimation in crowds: a survey

“…Is the purpose of the method to estimate the emotion of individuals within a crowd or the emotion of the crowd as a whole? Except for [11] The examined methods tend to model emotions in crowds either at the individual level (microscopic) or at a global (macroscopic) level. Microscopic models describe collectives at the individual level, modeling the emotions, behavior, actions and decisions of single crowd members.…”

“…The authors in [11] propose a hierarchical Bayesian model aiming to describe the crowd both at the microscopic and macroscopic level. This approach uses pedestrians trajectories to create a topological map by means of a self-organizing map (SOM), dividing the environment into zones.…”