Description:
Key Words:
Style Transfer,Machine Learning Applications,Possible Futures,World Building
Required Skills:
3D Modeling skills, Some lite coding may be required,
Required Software:
Rhino, Grasshopper, Twin motion or Unity, or Enscape
Required Hardware:
A capable computer, Webcam & Microphone,
Maximum number of participating students:
20
The process utilizes a machine learning style transfer technology made by the Oxford's Visual Geometry Group. We take the predetermined process and invert it. Thus we use a texture as the base image and an image as a texture, or in other cases we forgo the texture all together and instead opt to only feed the system base images; starving the system for a texture. This forces the system to produce uncanny, surreal, and possibly sublime results, outside the realm of expectation. These results can then be taken by the designer as a prompt, for further speculation and development based on a set of constraints. A more traditional modeling process follows, with the outcomes moving from "not possible to conceive" to a navigable experience.