Automatic tracking of multiply objects using GPUs

• Grant number: 11/12764-0
• Support Opportunities: Scholarships in Brazil - Scientific Initiation
• Start date: September 01, 2011
• End date: August 31, 2012
• Field of knowledge: Physical Sciences and Mathematics - Computer Science - Computing Methodologies and Techniques
• Principal Investigator: Yossi Zana
• Grantee: Antonio Henrique de Oliveira Fonseca
• Host Institution: Centro de Matemática, Computação e Cognição (CMCC). Universidade Federal do ABC (UFABC). Santo André , SP, Brazil

Abstract

Automatic visual tracking of objects is a recurrent theme in the literature, due to its computational complexity and importance of solving this problem. The potential gains of an automatic tracking are present in several areas, such as surveillance and experimental science. This project aims to develop an algorithm and its implementation with the capability to detect the trajectory of a large number of objects. The algorithm will use a novel combination of conventional methods. The implementation of the prototype system will be in Matlab and C/CUDA, taking advantage of the capacity for rapid development environment and high parallel processing power of GPUs. Preliminary tests with ants' video show high potential for success and significant scientific and technological potential contributions. The execution of this Scientific Initiation project ensures the candidate high lever scientific / technological academic preparation. (AU)