Joint regularized nearest points for image set based face recognition

作者：

Highlights：

• We proposed a greedy JRNP (GJRNP) without weight estimation needed in adaptive JRNP.

• GJRNP considers both collaboration and between-class discrimination of gallery sets.

• The query-set nearest points of GJRNP flexibly change for different gallery sets.

• Experiments on large-scale You Tube faces and 47 celebrities datasets are conducted.

• CMC curves, statistical significance test, and sensitivity analysis are added.

摘要

•We proposed a greedy JRNP (GJRNP) without weight estimation needed in adaptive JRNP.•GJRNP considers both collaboration and between-class discrimination of gallery sets.•The query-set nearest points of GJRNP flexibly change for different gallery sets.•Experiments on large-scale You Tube faces and 47 celebrities datasets are conducted.•CMC curves, statistical significance test, and sensitivity analysis are added.

论文关键词：Face recognition,Image set,Joint regularized nearest points,Sparse representation

论文评审过程：Received 18 September 2015, Revised 30 June 2016, Accepted 16 July 2016, Available online 6 August 2016, Version of Record 20 February 2017.

论文官网地址：https://doi.org/10.1016/j.imavis.2016.07.008