| "Rules: Logic and Applications" 2nd Workshop, Dec, 2019 |
| Aesthetic Morphisms |
| Jocelyn Ireson-Paine |
| www.jocelyns-cartoons.uk/rules2019/ |
| "Rules: Logic and Applications" 2nd Workshop, Dec, 2019 |
| Aesthetic Morphisms |
| Jocelyn Ireson-Paine |
| www.jocelyns-cartoons.uk/rules2019/ |
| Slides as PDF | Artistic Techniques Database Demo | Artistic Techniques Database Video | Contact |
|
I have argued against transformations that do the entire job by machine learning with no symbolic knowledge of aesthetics. However, I don't mind smaller "atomic" aesthetic transformations being done this way, provided they can be well specified in terms of an artist's goals.
On this and the next two slides, are a small proof-of-concept using Pix2Pix.
Pix2Pix: program for generic image-to-image transformation, using conditional adversarial networks. Originally by Phillip Isola et. al.
Examples: see Machine Learning for Artists's "Pix2Pix" page, and https://ml4a.github.io/guides/Pix2Pix/ and "Image-to-Image Demo: Interactive Image Translation with pix2pix-tensorflow" by Christopher Hesse. Try it online at the latter.
