Computational Design Laboratory

Jinmo Rhee’s “Deeprise” Project Receives Research Award

Title of Project: Deeprise

Award: Grand Prize in Digital Architecture & Architecture Startup Competition 2021 (DAAS 2021), Architectural Institute of Korea and Korea Ministry of Land, Infrastructure and Transport.

Date: 2021.12.02

Abstract:

Deeprise is a research prototype synthesizing a new high-rise building form based on the types of high-rise building form. Deeprise uses a deep learning (DL) method to investigate existing types of high-rise buildings and to generate new ones. The high-rise building form data is collected, comprehending diverse forms of high-rise building from major cities in the world to train a generative DL model to capture morphological features. After clustering the features, the deep learning model in the Deeprise can distinguish types of high-rise buildings and designers can use that information to generate novel high-rise building forms. This prototype demonstrates that generative DL models can uncover the latent types of architectural form in large datasets and can expand the typological interpretation of complex architectural forms. Besides, this prototype shows the potential of the proposed deep learning method for building massing design by developing a proposal of a high-rise building form based on three techniques: exploration, synthesis, and interpolation.

Author: Stella Shen
Category: News