Facial Feature Point Location in The Neural Network With Few Training Samples

DOI

10.2991/icmmse-16.2016.64How to use a DOI?

Keywords

Facial feature point location, Small sample training, Neural network, Sub-network

Abstract

This paper investigates a method to locate important facial feature points with small training samples. Firstly, the facial feature points are divided into several categories, then these various feature points are trained by LMBP neural network to get each sub-network. Outputs of these sub-networks can be combined to locate the important facial feature points. The experimental results show this kind of method based on neural network performs well, especially its calculation speed is fast, it can be applied in analysis of the facial expressions, facial reconstruction and other aspects.

Copyright

© 2016, the Authors. Published by Atlantis Press.

Open Access

This is an open access article distributed under the CC BY-NC license (http://creativecommons.org/licenses/by-nc/4.0/).

Download article (PDF)

TY  - CONF
AU  - Zhengyong Chen
AU  - Xiaohang Zhang
AU  - Jiaqi Shi
AU  - Shuang Zheng
AU  - Xiaodan Zou
PY  - 2016/03
DA  - 2016/03
TI  - Facial Feature Point Location in The Neural Network With Few Training Samples
BT  - Proceedings of the 2016 International Conference on Mechanics, Materials and Structural Engineering
PB  - Atlantis Press
SP  - 385
EP  - 390
SN  - 2352-5401
UR  - https://doi.org/10.2991/icmmse-16.2016.64
DO  - 10.2991/icmmse-16.2016.64
ID  - Chen2016/03
ER  -