id |
ecaade2023_183 |
authors |
Werker, Ines and Beneich, Kinza |
year |
2023 |
title |
Open AI in the Design Process - To what extent can text-to-image software support future architects in the early design process? |
source |
Dokonal, W, Hirschberg, U and Wurzer, G (eds.), Digital Design Reconsidered - Proceedings of the 41st Conference on Education and Research in Computer Aided Architectural Design in Europe (eCAADe 2023) - Volume 2, Graz, 20-22 September 2023, pp. 577–586 |
doi |
https://doi.org/10.52842/conf.ecaade.2023.2.577
|
summary |
The laborious creation of digital images could soon be a thing of the past. Text-to-image software generates images from text descriptions through artificial intelligence, the AI can map entirely new concepts and create images in a variety of artistic styles. Existing text-to-image software is already publicly available, but does it live up to its promise, and can it be more useful to architects in their search for inspiration than previous software that uses visual search to display images? In this paper, we address the opportunities and problems of text-to-image software. To answer our question, we use a key study, this is divided into two user groups. The subjects of group A are to use DALL·E 2 to search for inspiration for a design whose task is: Design a museum with a boat dock. The same design task is also given to the subjects of group B, with the difference that they are to use Pinterest to find inspiration.We will then contrast the results of these surveys. We will document the differences of the user experience and the output of DALL·E 2 to Pinterest as well as about advantages and disadvantages of DALL·E 2 and possible future developments, and application areas of text-to-image software. |
keywords |
text-to-image, DALL·E 2, Pinterest, early design process, picture generating, inspirational searching, AI |
series |
eCAADe |
email |
|
full text |
file.pdf (2,551,019 bytes) |
references |
Content-type: text/plain
|
last changed |
2023/12/10 10:49 |
|