FedNet: Federated Implementation of CNN for Facial Expression Recognition

In this project, we perform Facial Expression Recognition using a Federated Learning-based approach and demonstrate superior generalizability. The 5 ipynb files contain 5 different implementation of FedNet on Extended CK+ dataset and FER-2013 dataset. The .py file contains useful functions. The CK+ small dataset is a cropped subset of the original dataset.

---

Requirements

The project was carried out using TensorFlow 2.7.0, Scikit-Learn 1.0.2, opencv-python 4.5.5.

---

Citation

If you find this work useful in your research, please consider citing our paper:

@INPROCEEDINGS{siddiqui2022fednet,
  author={Siddiqui, Md. Saiful Bari and Shusmita, Sanjida Ali and Sabreen, Shareea and Alam, Md. Golam Rabiul},
  booktitle={2022 International Conference on Decision Aid Sciences and Applications (DASA)},
  title={FedNet: Federated Implementation of Neural Networks for Facial Expression Recognition},
  year={2022},
  volume={},
  number={},
  pages={82-87},
  doi={10.1109/DASA54658.2022.9765165}}

---

License

This project is licensed under the MIT License. See the LICENSE file for details.