Thanks for the 'easier' demo. I read the original readme and had a few goes at different pictures, with poor to abysmal results. After seeing the demo, and using the tool (with the readme open) I have gotten results far better than I thought I could.
I have still to find the limit of (my expectations of) the tool, but I am happier using it now.
For the next one, you need to be able to select a large part of the picture, click fix, and the optional neural reader will correct the picture to exactly what you had in mind. I can't see a problem with that - can you ?
Thanks
Paul