Modeling and synthesis of kinship patterns of facial expressions

作者：

Highlights：

• We generate realistic child videos from videos of the corresponding parents.

• Enhancing the training set with our generated data improves kinship verification.

• Our appearance network conserves the facial expressions while transforming identity.

• Action unit intensity leads to a better expression matching than facial landmarks.

摘要

•We generate realistic child videos from videos of the corresponding parents.•Enhancing the training set with our generated data improves kinship verification.•Our appearance network conserves the facial expressions while transforming identity.•Action unit intensity leads to a better expression matching than facial landmarks.

论文关键词：Kinship synthesis,Kinship verification,Temporal analysis,Facial action units,Facial dynamics

论文评审过程：Received 16 October 2017, Revised 12 May 2018, Accepted 12 September 2018, Available online 22 September 2018, Version of Record 9 October 2018.

论文官网地址：https://doi.org/10.1016/j.imavis.2018.09.012