Adaptive Global-Local Representation Learning and Selection for Cross-Domain Facial Expression Recognition

Implementation of papers:

• Adaptive Global-Local Representation Learning and Selection for Cross-Domain Facial Expression Recognition
  IEEE Transactions on Multimedia (IEEE TMM), 2024.
  Yuefang Gao, Yuhao Xie, Zeke Zexi Hu, Tianshui Chen, Liang Lin

Framework of AGLRLS.

Environment

Ubuntu 22.04.2 LTS, python 3.8.10, PyTorch 1.9.0.

Datasets

Application website： CK+, JAFFE, SFEW 2.0, FER2013, ExpW, RAF.

Six facial expression datasets.

Trained Models

The trained models are share in: baidu drive.

Usage

cd code
bash TrainOnSourceDomain.sh     # First step
bash TransferToTargetDomain.sh  # Second step

Citation

If you find our paper or code helpful, please cite our work.

@ARTICLE{10404024,
  author={Gao, Yuefang and Xie, Yuhao and Hu, Zeke Zexi and Chen, Tianshui and Lin, Liang},
  journal={IEEE Transactions on Multimedia}, 
  title={Adaptive Global-Local Representation Learning and Selection for Cross-Domain Facial Expression Recognition}, 
  year={2024},
  volume={26},
  number={},
  pages={6676-6688},
  keywords={Feature extraction;Adaptation models;Adversarial machine learning;Face recognition;Semantics;Data models;Representation learning;Domain adaptation;adverserial learning;Pseudo label generation;Facial expression recognition},
  doi={10.1109/TMM.2024.3355637}
}

Contributors

For any questions, feel free to open an issue or contact us:

• [email protected]