This is the official repo of the paper Confusing Image Quality Assessment: Towards Better Augmented Reality Experience:
@article{duan2022confusing,
title={Confusing image quality assessment: Towards better augmented reality experience},
author={Duan, Huiyu and Min, Xiongkuo and Zhu, Yucheng and Zhai, Guangtao and Yang, Xiaokang and Le Callet, Patrick},
journal={IEEE Transactions on Image Processing (TIP)},
year={2022}
}
The constructed CFIQA database and ARIQA database can be accessed using the links below.
The subjective data process codes can be found in code1_subjective_data_process.
The baseline traditional models can be found in code2_baselines.
The proposed CFIQA and ARIQA models can be found in code3_cfiqa_ariqa.
The metrics for measuring performance of IQA models can be found in code4_calculate_performance.
If you have any question, please contact huiyuduan@sjtu.edu.cn