D2S: Dynamic Distribution Supervision for Multi-Label Facial Expression Recognition

Lijuan Wang, Xin Zhang, Ning Jiang, Haiying Wu, Jufeng Yang

Abstract

Due to facial images usually evoking multiple emotions with different intensities, there exists much ambiguity in facial expression recognition (FER). Previous methods jointly optimize multi-label learning (MLL) and label distribution learning (LDL) to suppress ambiguity, which have achieved excellent performance. However, the different convergence speed of MLL and LDL make the model easily over-fitting. To address this problem, we propose a dynamic distribution supervision (D <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</sup> S) method, where the label distribution information is introduced as auxiliary supervision for multilabel classification. Specifically, we develop a multi-task framework in which MLL and LDL are optimized simultaneously. The losses are dynamically weighted to overcome the inconsistency of inter-losses optimization between the two tasks. Extensive experiments on the largest benchmark dataset, i.e., RAF-ML, demonstrate the superiority of the proposed method.
BibTeX
@inproceedings{wang2022d,
  author = {Lijuan Wang and Xin Zhang and Ning Jiang and Haiying Wu and Jufeng Yang},
  title = {{D<sup>2</sup>S: Dynamic Distribution Supervision for Multi-Label Facial Expression Recognition}},
  booktitle = {ICME},
  year = {2022},
  url = {https://openreview.net/forum?id=C2ceYJ8OcwN},
}

Loading related content...

Figure