Fusing landmark-based features at kernel level for face recognition

详细信息    查看全文

• 作者：Ke-Kun Huang^a ; ^{kkcocoon@163.com}Author Vitae ; Dao-Qing Dai^b ; ^{stsddq@mail.sysu.edu.cn}Author Vitae ; Chuan-Xian Ren^b ; ^{rchuanx@mail.sysu.edu.cn}Author Vitae ; Yu-Feng Yu^b ; ^{yuyufeng220@163.com}Author Vitae ; Zhao-Rong Lai^c ; ^{laizhr@mail2.sysu.edu.cn}Author Vitae
• 关键词：Face recognition ; Landmark-based feature ; Feature fusion ; Multi-kernel discriminant analysis ; Face identification ; Face verification
• 刊名：Pattern Recognition
• 出版年：2017
• 出版时间：March 2017
• 年：2017
• 卷：63
• 期：Complete
• 页码：406-415
• 全文大小：853 K
• 卷排序：63

文摘

An improved landmark-based multi-scale LBP feature is proposed. A novel model is proposed to fuse LBP feature and Gabor feature at kernel-level. An optimization algorithm is proposed to solve the kernel-level fusing model.