Progressive Prototype Evolving for Dual-Forgetting Mitigation in Non-Exemplar Online Continual Learning
Python 3.9.18
PyTorch 1.13.1+cu117
CIFAR-10 and CIFAR-100 will be automatically download. Download the MiniImageNet Datasets from this link.
# modify the --dataset_dir with your data path
bash r-cifar10.sh # Results on CIFAR-10
bash r-cifar100.sh # Results on CIFAR-100
bash r-mini.sh # Results on MiniImageNetThis project is mainly based on OnPro, GPM, and PCR.
If you find this work helpful, please cite:
@inproceedings{li2024progressive,
title={Progressive Prototype Evolving for Dual-Forgetting Mitigation in Non-Exemplar Online Continual Learning},
author={Li, Qiwei and Peng, Yuxin and Zhou, Jiahuan},
booktitle={Proceedings of the 32nd ACM International Conference on Multimedia},
pages={2477--2486},
year={2024}
}
Welcome to our Laboratory Homepage (OV3 Lab) for more information about our papers, source codes, and datasets.