Flexible-Context Activity Identification on Surveillance Cameras

• Grant number: 12/20738-1
• Support Opportunities: Scholarships in Brazil - Doctorate
• Start date: February 01, 2013
• End date: December 31, 2015
• Field of knowledge: Physical Sciences and Mathematics - Computer Science - Computing Methodologies and Techniques
• Principal Investigator: Hélio Pedrini
• Grantee: Marlon Fernandes de Alcantara
• Host Institution: Instituto de Computação (IC). Universidade Estadual de Campinas (UNICAMP). Campinas , SP, Brazil
• Associated research grant: 11/22749-8 - Challenges in exploratory visualization of multidimensional data: paradigms, scalability and applications, AP.TEM

Abstract

Surveillance systems are used in several application areas, whose specificities make modeling and implementation complex tasks. Surveillance systems can be classified regarding the use of technology or operation mode. Automatic action and activity identification is one of the alternatives to be used in surveillance systems. While the identification of actions seeks to determine subject movement patterns, the identification of activities aims at establishing a semantic linking a sequence of actions and interaction between individuals or objects. The techniques presented in the literature over the past five years use classifiers for motion and spatio-temporal classifiers for identifying actions and graphical models. Given this scenario, this paper proposes the development of an identification system of flexible activities to the context in which it is applied, using as a basis a system of identification of actions and a hierarchical statistical model-establishing the relationship between actions and activities. Results are quantitatively evaluated using public databases or by analyzing accuracy rate and response time of the system, and will be qualitatively evaluated by a system operator.

Scientific publications

(References retrieved automatically from Web of Science and SciELO through information on FAPESP grants and their corresponding numbers as mentioned in the publications by the authors)

DE ALCANTARA, MARLON FERNANDES; MOREIRA, THIERRY PINHEIRO; PEDRINI, HELIO; FLOREZ-REVUELTA, FRANCISCO. Action identification using a descriptor with autonomous fragments in a multilevel prediction scheme. SIGNAL IMAGE AND VIDEO PROCESSING, v. 11, n. 2, p. 325-332, FEB 2017. (11/22749-8, 12/20738-1)

Academic Publications

(References retrieved automatically from State of São Paulo Research Institutions)

ALCANTARA, Marlon Fernandes de. Human action identification in videos using descriptor with autonomous fragments and multilevel prediction. 2015. Doctoral Thesis - Universidade Estadual de Campinas (UNICAMP). Instituto de Computação Campinas, SP.