search

lidq92 / SFA

[official] No reference image quality assessment based Semantic Feature Aggregation, published in ACM MM 2017, TMM 2019
MIT License
80 stars 13 forks source link

Pytorch #1

Closed chungyau97 closed 4 years ago

chungyau97 commented 5 years ago

I can't really find a pytorch version of this code. Can anyone share with me? @lidq92

lidq92 commented 5 years ago

@chungyau97 This work is implemented using MATLAB and Caffe. Following README.md, you can easily use it to reproduce the work.

Of course, It is very easy to implement using Python and PyTorch. You are very welcome to follow the paper and this repository to implement by yourself. You may refer to some Python codes of VSFA, the extension to the VQA domain. You also need to implement PLSR with PLSRegression (from sklearn.cross_decomposition import PLSRegression). But we should note that the pre-trained model files between PyTorch and Caffe are different, and the performance may show a little variation.

When I am free, I may spend some time to reimplement this method with Python and PyTorch.

lidq92 commented 4 years ago

@chungyau97 Now you can have a look at SFA-pytorch