Português English
Contato

Proposta de Tese de Igor Rodrigues de Almeida


Detalhes do Evento


Aluno: Igor Rodrigues de Almeida
Orientador: Prof. Dr. Claudio Rosito Jung

Título: Crowd Analysis Using Local Neighborhood Coherence
Linha de Pesquisa: Processamento de Imagens e Visão Computacional e Reconhecimento de Padrões

Data: 13/07/2018
Horário: 10h
Local: Prédio 43412 – AUD-1 (auditório 1) do Instituto de Informática

Banca Examinadora:
– Prof. Dr. Marcelo Walter (UFRGS)
– Prof. Dr. Edson Prestes e Silva Junior (UFRGS)
– Prof. Dr. Luiz Gonzaga da Silveira Junior (UNISINOS)

Presidente da Banca: Prof. Dr. Claudio Rosito Jung

Abstract:  Many crowd analysis methods are developed in computer vision area. In this work we present an approach to explore characteristics inherent to human crowds – proxemics, and neighborhood relationship – for extract crowd features. We propose a crowd flow filtering approach for calibrated cameras that can be coupled to any generic optical flow method, and also a method to anomaly detection and localization using machine learning techniques. Both approaches were tested on publicly available datasets involving human crowded scenarios.

Keywords: Human Crowds. Computer Vision. Event Detection.