Feature fusion for facial landmark detection

作者：

Highlights：

• Novel generalized framework for combining facial landmark descriptors.

• The combination of landmark descriptors is an under-studied issue.

• Several feature fusion schemes for landmark detection are proposed and evaluated.

• The proposed approach offers dimensionality reduction and is easily extendable.

• The proposed approach can combine features extracted from 3D and 2D facial data.

摘要

Highlights•Novel generalized framework for combining facial landmark descriptors.•The combination of landmark descriptors is an under-studied issue.•Several feature fusion schemes for landmark detection are proposed and evaluated.•The proposed approach offers dimensionality reduction and is easily extendable.•The proposed approach can combine features extracted from 3D and 2D facial data.

论文关键词：Facial landmarks,Feature extraction,Feature fusion,Landmark detection

论文评审过程：Received 10 March 2013, Revised 18 September 2013, Accepted 8 March 2014, Available online 20 March 2014.

论文官网地址：https://doi.org/10.1016/j.patcog.2014.03.007