This code implements a deep learning model for predciting the satisfied user ratio (SUR) curve.
Run demo.ipynb to see how to extract deep features and train a deep model.
Please cite the following paper if you use the code:
@article{sur-featnet, title={SUR-FeatNet: Predicting the Satisfied User Ratio Curve for Image Compression with
Deep Feature Learning}, author={Lin, Hanhe and Hosu, Vlad and Fan, Chunling and Zhang, Yun and Mu, Yuchen and Hamzaoui,
Raouf and Saupe, Dietmar}, journal={Quality and User Experience}, year={2020}, volume={5}, number={1}, pages={5}}