zwx8981 / TSN-IQA

[TIP2024] Task-specific normalization for continual learning of blind image quality models

After nearly two and a half years of peer review, our work has finally been accepted by IEEE T-IP !

The codebase of Task-specific normalization for continual learning of blind image quality models

Requirement

torch 1.8+

torchvision

Python 3

scikit-learn

scipy

Usage

python BIQA_CL.py

(This work shares a similar codebase with BIQA_CL, where a more detailed instruction is available.)

Citation

Should you find this repo useful to your research, we sincerely appreciate it if you cite our paper 😊 :

@article{zhang2024tsniqa,
  title={Task-specific normalization for continual learning of blind image quality models},
  author={Zhang, Weixia and Ma, Kede and Zhai, Guangtao and Yang, Xiaokang},
  journal={IEEE Transactions on Image Processing},
  month={Mar.},
  volume={33},
  issue={},
  pages={1898 - 1910},
  year={2024}
}