> But edge detection is a lot easier on comics than on photos I would think That should be true. But deep learning based models work quite good on natural imag

> But edge detection is a lot easier on comics than on photos I would think That should be true. But deep learning based models work quite good on natural images as Vgg dataset is mostly natural. This is perhaps one-off usage/application on comics.