322 karma · joined November 25, 2013
However, I don't see any technical challenges in supporting Electron apps as well, other than constraints in RAM, performance and security of the Electron platform.
I know I am being a bit bitter for an open source project with less funding, but GIMP was supposed to be a poster boy GTK application in the first place.
While there are many benefits of end-to-end training, I don't think it might be best suited for this case. This is because we already know that the only useful feature in the document is the text, and not the contours or textures. Theefore wasting neurons in your neural network to learn the wastefulness of this features is just waste of resources. Furthermore, you benefit from even a larger corpus of learned data, which the char recognition engine has been trained on.