Enhancement of human body detection and tracking algorithm based on Viola and Jones framework

摘要

Computer vision application for the human body detection is a fundamental research problem in the areas such as surveillance, search and rescue missions, autonomous driving and other. Building an effective human body detector algorithm is a complex task due to articulation, clothing and illumination conditions. This paper evaluates human body detection and tracking methodology based on the Viola and Jones framework. Presented enhancement aims to address some of the drawbacks from their work which are, sensitivity to noise, poor performances with the complex backgrounds and limits on the scale and rotation.