A composite spatio-temporal modeling approach for age invariant face recognition
作者：
Highlights：
• A novel method is proposed that combines both texture and shape features.
• Face recognition models are built at different levels of data granularity.
• Experimentation is based on two well-known benchmarks, FG-NET and MORPH.
• Proposed method outperforms state of art recognition method on rank-1 accuracy.
• Proposed models support the simulation of aging effects at future time points.
摘要
•A novel method is proposed that combines both texture and shape features.•Face recognition models are built at different levels of data granularity.•Experimentation is based on two well-known benchmarks, FG-NET and MORPH.•Proposed method outperforms state of art recognition method on rank-1 accuracy.•Proposed models support the simulation of aging effects at future time points.
论文关键词：Anthropometric model,Local model,Personalized model,Integrated model,K nearest neighbor,Decision tree,Naive Bayes,Adaline Neural Network
论文评审过程：Received 6 June 2016, Revised 14 August 2016, Accepted 17 October 2016, Available online 29 October 2016, Version of Record 2 January 2017.
论文官网地址：https://doi.org/10.1016/j.eswa.2016.10.042
