Facial Point Detection Based on Multi-Modal Information and ASMS

摘要

Detecting a set of facial points is a crucial step for face recognition and facial expression analysis, yet the perfect facial point detector is yet to be developed. In this paper we present a method based on a combination of Multi-modal Information and Active Shape Model of Subsets to drastically reduce the time needed to search for a point's location and increase the accuracy and robustness of the algorithm. Using Multi-modal Information allows us to expand training samples and set up complete point distribution models (PDMs). On the other hand, training samples are grouped into different subsets, which makes detection of the points very fast and can make the algorithm robust to pose and illumination variations. In order to solve the local minima problem in fitting phase, Linear Programming Optimization (LPO) will be utilized. The proposed point detection algorithm was tested on WHU-3D-2D database, the results of which showed we outperform current state of the art point detectors.