Modeling urbanization patterns with generative adversarial networks

01/08/2018

∙

by   Adrian Albert, et al.

∙

0

∙

share

In this study we propose a new method to simulate hyper-realistic urban patterns using Generative Adversarial Networks trained with a global urban land-use inventory. We generated a synthetic urban "universe" that qualitatively reproduces the complex spatial organization observed in global urban patterns, while being able to quantitatively recover certain key high-level urban spatial metrics.