Architects often make confusing and cryptic statements about their work. To fix this, and make real the words of architects, I turned to off-the-shelf machine learning tools to create a pipeline for converting architect-speak into real images: the Real to Real Semantic Converter.

Designed @ Daniels
Toronto, Canada

Original Text:

“The Metropolis strives to reach a mythical point where the world is completely fabricated by man, so that it absolutely coincides with his desires.”

Rem Koolhaas

Machine Vision:

Real Meaning:

“A group of cows standing in the middle of a street.”

Photoreal Render:

Original Text:

“The house is a machine for living in.”

Le Corbusier

Machine Vision:

Real Meaning:

A kitchen with a refrigerator, stove, and sink.

Photoreal Render:

Original Text:

“... the German green uniforms made the place look gay and happy.”

Philip Johnson, speaking about Nuremberg rallies

Machine Vision:

Real Meaning:

A group of people standing next to each other.

Photoreal Render: