Deep Form Finding Using Variational Autoencoders for deep form finding of structural typologies

Jaime de Miguel*, Maria Eugenia Villafane, Luka Piskorec, Fernando Sancho-Caparrini

*Tämän työn vastaava kirjoittaja

Tutkimustuotos: Artikkeli kirjassa/konferenssijulkaisussaConference contributionScientificvertaisarvioitu

6 Lataukset (Pure)

Abstrakti

In this paper, we are aiming to present a methodology for generation, manipulation and form finding of structural typologies using variational autoencoders, a machine learning model based on neural networks. We are giving a detailed description of the neural network architecture used as well as the data representation based on the concept of a 3D-canvas with voxelized wireframes. In this 3D-canvas, the input geometry of the building typologies is represented through their connectivity map and subsequently augmented to increase the size of the training set. Our variational autoencoder model then learns a continuous latent distribution of the input data from which we can sample to generate new geometry instances, essentially hybrids of the initial input geometries. Finally, we present the results of these computational experiments and lay out the conclusions as well as outlook for future research in this field.

AlkuperäiskieliEnglanti
OtsikkoECAADE SIGRADI 2019: ARCHITECTURE IN THE AGE OF THE 4TH INDUSTRIAL REVOLUTION, VOL 1
ToimittajatJP Sousa, GC Henriques, JP Xavier
KustantajaeCAADe
Sivut71-80
Sivumäärä10
Vuosikerta1
ISBN (elektroninen)978-94-91207-17-4
TilaJulkaistu - 2019
OKM-julkaisutyyppiA4 Artikkeli konferenssijulkaisuussa
Tapahtuma37th Conference on Education and Research in Computer Aided Architectural Design in Europe (eCAADe) & 23rd Conference of the Iberoamerican Society Digital Graphics (SIGraDi) - Porto, Portugali
Kesto: 11 syyskuuta 201913 syyskuuta 2019

Julkaisusarja

NimieCAADe proceedings
ISSN (painettu)2684-1843

Conference

Conference37th Conference on Education and Research in Computer Aided Architectural Design in Europe (eCAADe) & 23rd Conference of the Iberoamerican Society Digital Graphics (SIGraDi)
MaaPortugali
KaupunkiPorto
Ajanjakso11/09/201913/09/2019

Siteeraa tätä