Low-quality facial biometric verification via dictionary-based random pooling
作者：
Highlights：
• Investigate face recognition in the realistic low-resolution surveillance world.
• Apply sparse-coding based high-quality dictionary into low-quality face recognition.
• Convert LR face recognition into a matching problem of HR visual words.
• Exploit random pooling for a better matching accuracy.
• Validate the proposed scheme in both theoretical and experimental worlds.
摘要
•Investigate face recognition in the realistic low-resolution surveillance world.•Apply sparse-coding based high-quality dictionary into low-quality face recognition.•Convert LR face recognition into a matching problem of HR visual words.•Exploit random pooling for a better matching accuracy.•Validate the proposed scheme in both theoretical and experimental worlds.
论文关键词：Facial biometrics,Low resolution,Sparse coding,Random pooling,Kernel LDA,Visual surveillance
论文评审过程：Received 17 February 2015, Revised 20 June 2015, Accepted 24 September 2015, Available online 9 November 2015, Version of Record 24 December 2015.
论文官网地址：https://doi.org/10.1016/j.patcog.2015.09.031
