Convolutional relation network for facial expression recognition in the wild with few-shot learning

作者：

Highlights：

• Recognize new emotion classes in the wild given only few examples from each.

• Salient discriminative feature learning and emotion similarity learning.

• Few-shot learning and transfer discriminative knowledge.

• Divide the emotion database into independent disjoint training and testing set.

摘要

•Recognize new emotion classes in the wild given only few examples from each.•Salient discriminative feature learning and emotion similarity learning.•Few-shot learning and transfer discriminative knowledge.•Divide the emotion database into independent disjoint training and testing set.

论文关键词：Facial expression recognition,Few-shot learning,Discriminative feature analysis,Feature learning

论文评审过程：Received 15 January 2020, Revised 31 July 2021, Accepted 4 October 2021, Available online 16 October 2021, Version of Record 1 November 2021.

论文官网地址：https://doi.org/10.1016/j.eswa.2021.116046