Visualizing my research using Kandinsky diagrams

I’ve always enjoyed art galleries and museums for their tranquility, and they’re really great places to get some valuable “thinking time”. I also know that I don’t really have a strong eye for art (although interpretations can clearly be subjective). I’ve often gotten a lot out of visual abstracts for papers, and I’ve been curious about AI generated art (e.g. DALLE), so naturally I was wondering whether this could be a relatively simple task to generate an “art gallery” of research I’ve done.

In this spirit I found the works of Wassily Kandinsky really fascinating, because they have really straightforward and simple ingredients (e.g. sine waves, rectangles, circles) that are oriented in different ways. It also happened to have a very nice R package. Its certainly less sophisticated that currently derived machine-learning methods (like variational autoencoders or DALLE), and the downside is that it only generates art in the style of Kandinsky, rather than mimicing the stylings of another artist. The upside is that with a couple of scripts, you too can generate your own gallery based on your abstract!

1 / 8
2 / 8
3 / 8
4 / 8
5 / 8
6 / 8
7 / 8
8 / 8

Lauterber et al 2022
Biddanda et al 2022
Zhang et al 2022
Biddanda et al 2022
Biddanda et al 2020
Damgaard et al 2018
Chiang et al 2018
Waldman et al 2016