Yifan-Feng-CS / FCIL

This is the formal code implementation of the CVPR 2022 paper 'Federated Class Incremental Learning'.

Official Pytorch Implementation for GLFC

[CVPR-2022] Federated Class-Incremental Learning

This is the implementation code of the CVPR 2022 paper "Federated Class-Incremental Learning".

You can also find the arXiv version with supplementary materials here. More related works are provided at Dynamic Federated Learning, please work with us to make FL more practical and realistic.

Framework:

Prerequisites:

* python == 3.6
* torch == 1.2.0
* numpy
* PIL
* torchvision == 0.4.0
* cv2
* scipy == 1.5.2
* sklearn == 0.24.1

Datasets:

• CIFAR100: You don't need to do anything before running the experiments on CIFAR100 dataset.

• Imagenet-Subset (Mini-Imagenet): Please manually download the on Imagenet-Subset (Mini-Imagenet) dataset from the official websites, and place it in './train'.

• Tiny-Imagenet: Please manually download the on Tiny-Imagenet dataset from the official websites, and place it in './tiny-imagenet-200'.

Training:

• Please check the detailed arguments in './src/option.py'.

python fl_main.py

Performance:

• Experiments on CIFAR100 dataset

• Experiments on Imagenet-Subset (Mini-Imagenet) dataset

Related Works

We apply federated class-incremental learning to semantic segmentation task.

1. [CVPR-2023] Federated Incremental Semantic Segmentation [Code]

Citation:

If you find this code is useful to your research, please consider to cite our paper.

@InProceedings{dong2022federated,
    author = {Dong, Jiahua and Wang, Lixu and Fang, Zhen and Sun, Gan and Xu, Shichao and Wang, Xiao and Zhu, Qi},
    title = {Federated Class-Incremental Learning},
    booktitle = {IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)},
    month = {June},
    year = {2022},
}

Contact:

• Lixu Wang: [e-mail]
• Jiahua Dong: [e-mail]