A joint learning method with consistency-aware for low-resolution facial expression recognition

Yuanlun Xie, Wenhong Tian, Liang Song, Ruini Xue, Zhiyuan Zha, Bihan Wen

Abstract: Highlights•Design a joint learning framework to alleviate the low-resolution FER challenge.•Design novel attention-aware consistency and prediction consistency losses.•Extensive experiments demonstrate the effectiveness of the proposed solution.