Comparação de técnicas de rastreamento de objetos

Abstract

Tracking is a field of study that has been growing drastically in recent years due to the great interest in the fields of applications. Consequently, several algorithms, methods and theories were developed and proposed in order to track objects on the scene. The objective of the work is to implement, analyze and compare some of the existing tracking techniques in diferente scenarios. The comparison was performed between two classic methods and a technique that uses the concepts of machine learning, considered the state-of-the-art in this area. Both visual and numerical results are presented for each method.