color palettes are essential to any type of kind of visual or graphic design. A designer has to determine a handful of essential colours to make a style work, making phone calls on what’s eye catching or what sets the mood appropriately. one of the issues is that it relies heavily on subjective judgement, rather than any type of understood mathematical formula. There are guidelines one can apply, however guidelines can likewise be creatively broken, so it’s never a easy task. To this end, [Jack Qiao] produced colormind.io, a tool that utilizes neural nets to generate color palettes.
It’s a fun tool – there’s a choice of palettes produced from prominent media as well as sunset photos, in addition to the choice to produce customized palettes yourself. Colours can be locked so you can iterate around those you like, discovering others that match well. The results are outstanding – the tool is able to generate palettes that seem to blend rather well. We were not able to force it to produce anything really garish in spite of a few attempts!
The blog discusses the software application behind the curtain. After very first experimenting with a type of neural web understood as an LSTM, [Jack] discovered the results as well bland. The network was scared to be wrong, so would select values extremely much “in the middle”, leading to soft palettes of browns as well as greys. After changing to a less accuracy-focused network understood as a GAN, the results were much better – [Jack] states the network now produces what it believes to be “plausible” palettes. The code has been uploaded to GitHub if you’d like to play around with it yourself.
Check out this primer on neural nets if you’d like to discover more. We’d like to understand – exactly how do you pick a combination when starting a project? let us understand in the comments.