A WALK IN LATENT CITY

MSCD | spring 22'

PROJECT IN EARLY DEVELOPMENT

PROJECT IN EARLY DEVELOPMENT

awilc city.PNG

In the realm of architecture and urban design, to what extent can an understanding of machine learning’s latent space provide a new understanding of context and space within the design process? And through what means is that achieved that responds to a designer’s intuitive approach?

 

My project’s objective is to develop a  three-dimensionalized experience utilizing the 3D model and machine learning model Rhee has currently been developing and interfacing with. My proposal to the research question is that a first-person situation, whether it is VR or simply first-person control, interacts with a designer’s more cognitive understanding of space and shape. The goal is to provide a link between a 3D representation of Montreal’s urban landscape and that of 3D latent space of building form data Jinmo Rhee has developed, all under one singular interactive experience.

comp1.PNG

ARCHITECTURE'S URBAN MAPPING VS ITS LATENT SPACE MAPPING

comp2.PNG

JINMO RHEE'S CURRENT 2D INTERFACE OF HIS VAE

comp3.PNG

GAMEPLAY INTERFACE EXPLORING TWO SPACES