I love when they can take algorithms like this and turn their output into something tangible.
The next step I’d see in this software is giving the software images of actual people to have it analyze their faces as the inputs for the shape and features of face to generate in the painting.
I want to be in a Rembot.
Most of what makes this work, is the age of the source material. I’m trying to imagine a computer generated Frank Lloyd Wright building, or a Rolling Stones song, or a Dale Chihuley. Any statistical average of typical works, is not going to be a new work. I’d love to be wrong about this.
Statically, you could create a Jesus everyone would believe in.
I’m fairly certain this was actually done using a reel of Müller-Fokker tape. John Sladek would be proud or amazed or shocked at this, much as am I.
This topic was automatically closed after 5 days. New replies are no longer allowed.
