Paradigmatic Landscapes

Posted by Yann Le Guennec | Mon Aug 17th 2009 12:15 p.m.

Each time you access a picture from the "Paradigmatic landscapes"
project, 3 portions, from 3 pictures selected randomly in a defined set
of pictures, are extracted, resized and superposed to compose a new
picture. Thus, each generated landscape is different from another. There
is a poor probability to generate 2 identical pictures. "Paradigmatic
landscapes" is declined into several series, involving different sets of
defined pictures serving as bases for the compositions. "Paradigmatic
landscapes" project explores the "if" and the "how" a very simple
algorithm and sets of pictures (1, 2, 3) can generate some kinds of
visual patterns and exceptions, according to some ideas about some kinds
of landscapes, but without analyzing further any data model or trying to
explicit what happen when generating and looking at (references to)
landscapes.

http://www.datapainting.com/paradigmatic-landscapes/
  • River Hunt | Thu Aug 20th 2009 5:58 p.m.
    Hey, thanks for sharing, that's a pretty cool idea to work with.

    I'm really interested in the concept of the observed being changed by the process of observing. Quantum Mechanical art. It would be really interesting to produce some code that demonstrates a level of A.I, and is able to learn from the interaction with the user (positional clicks or whatever) as to how to merge/modify the images, to produce images that through various iterations come to reflect the consciousness of the user in some way.

    River.

Your Reply