Emotion, personality and cultural aspects in crowds : towards a geometrical mind

Autor: Favaretto, Rodolfo Migon
Přispěvatelé: Musse, Soraia Raupp
Jazyk: angličtina
Rok vydání: 2019
Předmět:
Zdroj: Biblioteca Digital de Teses e Dissertações da PUC_RS
Pontifícia Universidade Católica do Rio Grande do Sul (PUCRS)
instacron:PUC_RS
Popis: The study of human behavior is a subject of great scientific interest and probably an inexhaustible source of research. The analysis of pedestrians and groups in crowds is an object of interest in several areas of application, such as security, entertainment, planning of environments in public spaces and social sciences. Cultural and personality aspects are attributes that influence personal behavior and affect the group to which the individual belongs. In this sense, the present thesis discusses different ways to characterize individuals and groups in crowds, with the purpose of proposing a computational model to extract pedestrian characteristics in video sequences. The proposed model considers a series of characteristics of the pedestrians and the crowd, such as number and size of groups, distances, speeds, among others, and performs the mapping of these characteristics in personalities, emotions and cultural aspects, considering the Cultural Dimensions of Hofstede (HCD), the Big-Five Personality Model (OCEAN) and the OCC Emotional Model. The main hypothesis is that there is a relationship between so-called intrinsic human variables (such as emotion) and the way people behave in space and time. As one of the main contributions, four large dimensions of geometric characteristics (Big4GD) were proposed: I - Physical, II - Personal and Emotional, III - Social and IV - Cultural, which seek to describe the behavior of pedestrians and groups in the crowd. The GeoMind tool was developed for the purpose of detecting the four geometric dimensions from video sequences. In addition, several analyzes were carried out with the purpose of validating the proposed model, from comparing results with the literature, including the comparison of spontaneous multitudes from several countries and controlled experiments involving Fundamental Diagrams. O estudo do comportamento humano ? um tema de grande interesse cient?fico e, provavelmente, uma fonte inesgot?vel de pesquisa. A an?lise de pedestres e grupos em multid?es ? objeto de interesse em diversas ?reas de aplica??o, tais como seguran?a, entretenimento, planejamento de ambientes em espa?os p?blicos e ci?ncias sociais. Aspectos culturais e de personalidade s?o atributos que influenciam o comportamento pessoal e afetam o grupo ao qual o indiv?duo pertence. Neste sentido, a presente tese discute diferentes maneiras de caracterizar indiv?duos e grupos em multid?es, com o objetivo de propor um modelo computacional para extrair caracter?sticas de pedestres em sequ?ncias de v?deos. O modelo proposto considera uma s?rie de caracter?sticas dos pedestres e da multid?o, tais como quantidade e tamanho de grupos, dist?ncias, velocidades, entre outros e realiza o mapeamento destas caracter?sticas em personalidades, emo??es e aspectos culturais, considerando as Dimens?es Culturais de Hofstede (HCD), o modelo de personalidades Big-Five (OCEAN) e o modelo de emo??es OCC. A principal hip?tese ? que existe rela??o entre vari?veis ditas intr?nsecas aos humanos (como emo??o) e a maneira pela qual as pessoas se comportam no espa?o e no tempo. Como uma das principais contribui??es, foram propostas quatro grandes dimens?es de caracter?sticas geom?tricas (Big4GD): I - F?sica, II - Pessoal e Emocional, III - Social e IV - Cultural, as quais buscam descrever o comportamento dos pedestres e grupos na multid?o. A ferramenta GeoMind foi desenvolvida com o prop?sito de detectar as quatro dimens?es geom?tricas, a partir de sequ?ncias de v?deos. Al?m disso, diversas an?lises foram realizadas com o intuito de validar o modelo proposto, desde confrontando resultados com a literatura, incluindo a compara??o de multid?es espont?neas de diversos pa?ses e experimentos controlados envolvendo Diagramas Fundamentais.
Databáze: OpenAIRE
Externí odkaz: