search

csxmli2016 / textbsr

This is a simple text image blind super-resolution model, using BSRGAN
Other
87 stars 5 forks source link

Can provide x1 and x2 pre-trained model? #1

Open Ken1256 opened 1 year ago

Ken1256 commented 1 year ago

Can provide x1 and x2 pre-trained model?

Thanks

csxmli2016 commented 1 year ago

Can provide x1 and x2 pre-trained model?

Thanks

Hi, this model is only a simple baseline (ESRGAN) of our CVPR23 paper (MARCONet) for practical usage. x1 and x2 model may have lower priority. If other people also have a need, I will focus on training them.

HuaqingHe commented 1 year ago

Hi~ This is a great job, thank you for such a great integration code. I have the same requirement as above. In real usage scenarios, the pictures are relatively large, and the resolution of X4 often leads to a series of problems such as the inference is too slow, the memory bursts. So can you announce a X2 or X1 model?

csxmli2016 commented 1 year ago

Sure. I will train new models for x1 and x2 scenarios. Currently, I have other tasks to do, so they may have lower priority.