Multi-Kernel Appearance Model

作者：

Highlights：

• Multi-Kernel Appearance Model is a new facial point detector.

• Multi-Kernel SVM combines multi-resolution features.

• A SVM cascade combines increasingly complex kernels to reduce the computational time.

• A shape model fitting introduces constraints between SVM detections.

摘要

•Multi-Kernel Appearance Model is a new facial point detector.•Multi-Kernel SVM combines multi-resolution features.•A SVM cascade combines increasingly complex kernels to reduce the computational time.•A shape model fitting introduces constraints between SVM detections.

论文关键词：Facial feature localization,Multiple-kernel learning,Two-stage classifiers,SIFT descriptor,Deformable model alignment,Gauss–Newton optimization

论文评审过程：Received 23 May 2012, Revised 2 April 2013, Accepted 25 April 2013, Available online 11 May 2013.

论文官网地址：https://doi.org/10.1016/j.imavis.2013.04.006