Author(s)

Shiyi Liu

Corresponding Author

Shiyi Liu

Abstract

In this paper, we propose an analysis of some methods for facial landmark detection (FLD). Recently, facial landmark detection has become a popular topic due to its importance in computer vision area. There has been some novel methods achieving breakthrough in different challenges such as accuracy improvement, detection with severe occlusions and large head poses. These FLD are majorly based on regression process. However, they have their own characteristics in the algorithm by adopting diverse approaches to optimize the regression such as deep learning, Gaussian process, local binary features and so nonexperimental results show all of them can achieve improvements compare to state-of-art methods. With facial point detection, information of human face can be utilized for facial analysis applications, such as face recognition, face synthesis and age estimate.

Keywords

Component, facial alignment, regression, deep learning.