Deep discriminant generation-shared feature learning for image-based kinship verification

作者：

Highlights：

• We propose a new deep discriminant generation-shared feature learning for kinship verification.

• We design a cross-generation difference loss to relieve divergence between different generations.

• We design an intra-generation discriminant loss to improve the discriminability of features.

• Our approach achieves comparable or better performance comparing to the state-of-the-art methods.

摘要

•We propose a new deep discriminant generation-shared feature learning for kinship verification.•We design a cross-generation difference loss to relieve divergence between different generations.•We design an intra-generation discriminant loss to improve the discriminability of features.•Our approach achieves comparable or better performance comparing to the state-of-the-art methods.

论文关键词：Kinship verification,Face recognition,Feature learning,Cross generation

论文评审过程：Received 24 May 2021, Revised 24 September 2021, Accepted 30 October 2021, Available online 15 November 2021, Version of Record 20 November 2021.

论文官网地址：https://doi.org/10.1016/j.image.2021.116543