When a computer science major becomes an artist

Perspective is hard, and redrawing the same design from different angles is frustrating.
*attempts to learn perspective*
*gives up and learns 3D modelling instead*

These 3D renders are great for perspective and different angles but they're not very pretty.
*attempts to learn digital painting*
*gives up and uses neural networks to process the renders*

These neural networks make everything pretty but it's a pain uploading the render, and I worry the site/app might disappear.
*tries digital painting again*
*gives up and learns how to run and train my own neural networks*

I mean I haven't actually installed the neural network yet, because that would require getting out of bed and installing things on the Windows machine, which I'm not awake enough for today. But it looks like fun :D

EDIT: I am going to be installing packages until I die.

A vague attempt at documenting my attempt to make neural network art

Neural Network/Deep Learning/Style Transfer art is this sort of thing.

This is mostly for my own reference when I next feel up to poking at the problem some more.

Unfortunately, this stuff is still fairly cutting edge, and all the documentation assumes you're an AI researcher running Linux. I'm sure there'll be simple to use graphics programs aimed at artists eventually, but for now the only things along those lines are websites and apps, which lack the flexibility I want.
( Read more... )