Pretrained model for target model

[lyumingzhi]

Hi! Thank you very much for your excellent work. Would you mind providing the fine tune model of the work "Split then Refine: Stacked Attention-guided ResUNets for Blind Single Image Visible Watermark Removal."? Since i find the model can not work on clean customized input, which is not suitable for attack experiment. May I know how you do the experiment of attacking this model, especially how to prepare the model?

[thinwayliu]

This is the model I trained myself, I hope it will help you.
https://pan.baidu.com/s/1e1IO26Il5yvg31rhXFKc2Q Code: 5yta