Video-based face classification approach: A survey

Abstract

The goal of this paper is to provide; an up-to-date, critical and comprehensive; survey of existing literature on face classification. The survey of this research paper includes the research work ranges from single: frame, object, classifier to multi: frame, object, classifier approach that provides robust and efficient solution for several applications in the area of biometrics, personal security, law enforcement, information security, entertainment, smart card and access management. Such applications have several constraints in terms of complexity of processing requirements and thus present a wide range of technical challenges. The detail overview of techniques for segmentation/location of the face, tracking, feature extraction and recognition are reviewed. As human face is a dynamic object having high degree of variability in its appearance, that makes face detection a difficult problem in computer vision. In this paper we will provide a brief introduction, different approaches and related state of the art recent work in the field of face recognition and classification specially for videos like global transform and feature based methods. At the end critical analysis based on literature results and conclusion have been provided.