A Comprehensive Survey of Algorithms for Face Tracking in different Background Video Sequence

Abstract

Video processing is an interesting research zone in image processing. Face tracking is part of video processing, where the face regions need to be detected and tracked. In this paper, we present a survey of some of the familiar algorithms that are used for tracking the face(s) in different background challenging video sequences. Mean-Shift is an important algorithm that is based on the displacement of points. Improvisation of Mean-Shift lead to the development of CAMSHIFT; the latter is one of the robust chromatic tracking approach developed till date. KLT is an efficient point tracking algorithm. This paper also includes a survey of different motion estimation algorithms, which are classified as either pixel based or feature based. At the end, recent developments help in knowing the relevant works that are being carried out now a days.