Metric Learning Based Multi-Patch Ensemble for High Precision Face Verification

DOI

10.2991/ceis-16.2016.32How to use a DOI?

Keywords

face recognition; deep learning; CNN; LFW; metric learning

Abstract

Face verification under video surveillance is an important issue in computer vision for decades. Several methods on automatic face verification have significantly raised the accuracy by using Gabor wavelets, high-dimensional LBP features, Fisher vectors, Joint Bayesian, etc. Especially the usage of Deep Convolution Neural Networks(CNNs), artificial intelligence beats humans for the first time on Labeled Faces in the Wild(LFW) face verification task. In order to further improve the verification accuracy, we propose an approach that combines a multi-patch deep CNN by using two step metric learning. Experiments show that our method archives an accuracy of 99.37% on LFW, and 92.80% on YouTube Faces(YTF), which is very competitive with the state-of-the-art.

Copyright

© 2017, 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  - Dang-Dang Chen
AU  - Lan-Qing He
AU  - Zhong-Dao Wang
AU  - Sheng-Jin Wang
PY  - 2016/11
DA  - 2016/11
TI  - Metric Learning Based Multi-Patch Ensemble for High Precision Face Verification
BT  - Proceedings of the 2016 International Conference on Computer Engineering and Information Systems
PB  - Atlantis Press
SP  - 164
EP  - 168
SN  - 2352-538X
UR  - https://doi.org/10.2991/ceis-16.2016.32
DO  - 10.2991/ceis-16.2016.32
ID  - Chen2016/11
ER  -