Variance-guided attention-based twin deep network for cross-spectral periocular recognition
作者：
Highlights：
• Cross-spectral matching reduces the gap amid periocular images in different spectra.
• Attention module focuses on the significant regions of the heterogeneous input.
• Variance-guided objective function accounts for both probability and feature space.
• Intermediate layer visualizations and ablation studies show efficacy of VGACNet.
摘要
•Cross-spectral matching reduces the gap amid periocular images in different spectra.•Attention module focuses on the significant regions of the heterogeneous input.•Variance-guided objective function accounts for both probability and feature space.•Intermediate layer visualizations and ablation studies show efficacy of VGACNet.
论文关键词：Attention,Cross-spectral,Convolutional neural network,Heterogeneous,Periocular
论文评审过程：Received 25 December 2019, Revised 1 August 2020, Accepted 31 August 2020, Available online 8 September 2020, Version of Record 22 September 2020.
论文官网地址：https://doi.org/10.1016/j.imavis.2020.104016
