PixelArt is now Live!

By Samuel Muiruri | March 4, 2019 | Django


This is a project I've been working on for a while to give some context to it, here's a couple of android apps that can do something similar

App name: Color Splash

There's more than one app with this name so the best way to know it is by its icon.

You upload an image, it first turns it to grayscale and then you choose which parts of the image you want to retain color by brushing with your finger on top of the image. You can also change the color by selecting the new color and still using your finger to color, it's hard to be precise though on where you want the changes to apply.

App name: Color Changer

App Icon:

Works the same way, though the UI feels a lot better and more intuitive.

App name: Color Splash

App Icon: 

You still use it the same way but by far compared to the other two has the most user-friendly interface.

In contrast when I designed the part of PixelArt that would handle this feature what I had in mind was something to be a challenge to what Machine Learning was doing with coloring images, basically find out the process of coloring an image manually and see how this insight could be reverse engineered as I learnt Machine Learning to coloring images.

Here's an example from an actual demo screen recorded and available on youtube using PixelArt:

Before and After Pictures

I already decided to make the code available on github and open source so everyone can use it. The youtube videos was taking it a step further so I can show you how to set it up locally on you machine. Finally I've been working on putting this up live and I've finally done it and you can play with it here, if your curious how to use it, just watch the youtube playlist as it covers the basics on how it works.