Abstract
AbstractThis article highlights work using machine learning in collaboration with designers for speculative world building. The process is unique because of the feedback loop, between the designer and the computational process. Worldbuilding is a speculative practice and requires vision and courage on the part of the designer. Working with machine learning neural style transfer (NST) allows the designers to consider possibilities humanity may not otherwise allow ourselves to imagine. This is important because human imagination paves the path for the future of humankind. Imagining a sustainable future requires considering unconventional solutions. Imagining non-probable futures allows humanity to glean desirable aspects to strive for. Even if a conceived future is impossible within the built environment, there are many opportunities for people to inhabit these environments virtually. Letting yourself get lost in these places is a form of travel, even when conditions limit one's ability to physically do so.
Publisher
Springer Nature Singapore
Reference19 articles.
1. Bratton BH, Stack T (2016) On software and sovereignty. The MIT Press, Cambridge, MA, p 213
2. Cook P (2020) Instant city in a field, typical set-up. Archigram Archives, Archigram. https://www.archigram.net/portfolio.html
3. Gatys LA, Ecker AS, Bethge M (2016) Image style transfer using convolutional neural networks. In: 2016 IEEE conference on computer vision and pattern recognition (CVPR), pp 2414–2423. https://doi.org/10.1109/CVPR.2016.265
4. Hassan MU (2021) Evolution of style transfer techniques. VGG16-convolutional network for classification and detection, 24 Feb 2021. https://www.neurohive.io/en/popular-networks/vgg16/
5. Kant I, Bernard JH (1914) Kant's critique of judgement. Macmillan, London