[dalerus]

Wow, that's an amazing level of dedication to mapping this all out. Looking at the about page, it looks like an good exercise in machine learning.

The developer talks about the process of generating the test data here: http://www.danvk.org/wp/2013-04-20/generating-training-data/

[Lanzaa]

This looks like a very similar problem to the one dropbox wrote about in "Fast and Accurate Document Detection for Scanning": https://blogs.dropbox.com/tech/2016/08/fast-and-accurate-doc... https://news.ycombinator.com/item?id=12254517