Orthogonal neighborhood preserving discriminant analysis with patch embedding for face recognition
作者：
Highlights：
• Intra-class compactness, intra-class structure, and inter-class separability are together integrated into patch embedding.
• The intra-class structure information is organically merged into the intra-class compactness loss.
• A fast-orthogonal strategy is introduced to obtain a projection matrix with orthogonal columns.
• The kernel ONPDA performed in reproducing kernel Hilbert space is employed to induce nonlinear maps.
摘要
•Intra-class compactness, intra-class structure, and inter-class separability are together integrated into patch embedding.•The intra-class structure information is organically merged into the intra-class compactness loss.•A fast-orthogonal strategy is introduced to obtain a projection matrix with orthogonal columns.•The kernel ONPDA performed in reproducing kernel Hilbert space is employed to induce nonlinear maps.
论文关键词：Neighborhood preserving,Inter-class separability,Orthogonal projection,Patch alignment,Face recognition
论文评审过程：Received 6 May 2019, Revised 21 March 2020, Accepted 13 May 2020, Available online 16 May 2020, Version of Record 29 May 2020.
论文官网地址：https://doi.org/10.1016/j.patcog.2020.107450
