TASS - Official PyTorch Implementation

[CVPR 2024] Task-Adaptive Saliency Guidance for Exemplar-free Class Incremental Learning

Paper

Usage

• Training on CIFAR-100 dataset:

$ python main.py

Low-level Map

You can download the pre-computed low-level map of CIFAR100 and Tiny-ImageNet for your experiments. For better usage, we upload the code for processing low-level map in DFI/, please follow DFI's instruction to prepare the model and images.

Citation

If you use this code for your research, please consider citing:

@article{liutask,
  title={Task-Adaptive Saliency Guidance for Exemplar-free Class Incremental Learning},
  author={Liu, Xialei and Zhai, Jiang-Tian and Bagdanov, Andrew D and Li, Ke and Cheng, Ming-Ming}
}