Virtual dictionary based kernel sparse representation for face recognition
作者：
Highlights：
• KCDVD can automatically yield virtual dictionary used to represent the samples.
• KCDVD can effectively address the undersampling problem in face recognition.
• KCDVD exploits the coordinate descent scheme to solve the representation models.
• KCDVD is easy to implement and is much faster than other similar methods.
• KCDVD outperforms many state-of-the-art classification methods.
摘要
•KCDVD can automatically yield virtual dictionary used to represent the samples.•KCDVD can effectively address the undersampling problem in face recognition.•KCDVD exploits the coordinate descent scheme to solve the representation models.•KCDVD is easy to implement and is much faster than other similar methods.•KCDVD outperforms many state-of-the-art classification methods.
论文关键词：Kernel sparse representation for classification (KSRC),Virtual dictionary,Coordinate descend,Face recognition
论文评审过程：Received 3 June 2016, Revised 30 September 2017, Accepted 3 October 2017, Available online 14 October 2017, Version of Record 31 October 2017.
论文官网地址：https://doi.org/10.1016/j.patcog.2017.10.001
