The Manga Me app instantly transforms your photo into a Japanese style manga/anime character. You will be able to create an awesome new profile pic or the more ambitious could even make a manga or visual novel all on your own without learning how to draw.
this is an interesting idea, but it's worth mentioning that our current understanding of neural networks is partly a black box. when using these algorithms in completely new fields you never know what you're going to get until you see the result. my point is that the success of your project depends mainly on something you have little control over and is hard to predict. i've heard of a few pix2pix examples that failed their initial intentions.
also, current pix2pix projects do a 1-to-1 mapping regarding the locality of the image information. take the cat implementation for example. precisely where you draw a line the algorithm will fit an edge of the generated cat. your algorithm needs to learn that the shape of a face may be changed to simulate the manga drawing style. i'm not saying this isn't possible, but as far as i know it has never been done with pix2pix.
may i ask what your background is regarding deep learning? are you just using pix2pix or do you deeply understand what its doing under the hood?
@gopietz Having used pix2pix I think this will be very difficult. Pix2pix is great for texture generation but bad at creating structure, like in the photo->map example straight lines and arcs tend to come out looking "liquified"
Manga Me
Huemint