Streetscape augmentation using generative adversarial networks: Insights related to health and wellbeing

Abstract

Deep learning using neural networks has provided advances in image style transfer, merging the content of one image (e.g., a photo) with the style of another (e.g., a painting). Our research shows this concept can be extended to analyse the design of streetscapes in relation to health and wellbeing outcomes. An Australian population health survey (n=34,000) was used to identify the spatial distribution of health and wellbeing outcomes, including general health and social capital. For each outcome, the most and least desirable locations formed two domains. Streetscape design was sampled using around 80,000 Google Street View images per domain. Generative adversarial networks translated these images from one domain to the other, preserving the main structure of the input image, but transforming the 'style' from locations where self-reported health was bad to locations where it was good. These translations indicate that areas in Melbourne with good general health are characterised by sufficient green space and compactness of the urban environment, whilst streetscape imagery related to high social capital contained more and wider footpaths, fewer fences and more grass. Beyond identifying relationships, the method is a first step towards computer-generated design interventions that have the potential to improve population health and wellbeing.

The “Paris-End” of Town? Deriving Urban Typologies Using Three Imagery Types

“…To better capture the street and building structure, pre-processing could be applied (edge detection, mean shifts, or other computer vision techniques) to force a stronger emphasis on the urban structure rather than the details. Further additional work could be performed, such as deconstructing and reconstructing imagery, removing features (such as automotive traffic, leaving only structures), and, for targeted outcomes (such as health and social capital), using generative adversarial networks [58], enabling comparative hypothetical typology scenarios.…”

Going Green and Going Dense: A Systematic Review of Compatibilities and Conflicts in Urban Research

“…This is the case of two studies conducted in Turku, Finland, that assessed the impact of the built environment, namely children's possibilities for independent mobility and opportunities to access environmental affordances, finding that moderate urban densities associated with the presence of green spaces seem to have childfriendly characteristics by both promoting independent access to meaningful places and good perceived health [81] or the likelihood of liking the place [82]. Additionally, a study conducted in Melbourne exploring differences in streetscape imagery from areas with good and bad perceived health and well-being outcomes, revealed that areas with higher self-reported health are characterized by both sufficient green space and compactness of the urban environment [83]. Similarly, in a study based on a household survey and geographical data in Guangzhou, China, Qiu, et al [84] found that both building density and per capita green area are positively correlated with mental health.…”

Modeling and interpreting road geometry from a driver's perspective using variational autoencoders

“…Advanced methodologies based on machine learning enable understanding the inherent structure of complex and large-scale data. Convolutional neural networks have been used to solve complex computer vision tasks, such as feature extraction, image generation, object detection, and classification (Antoniades et al, 2018;Dai et al, 2019;Hua, Wang, Lu, Liu, & Khalid, 2019;Krizhevsky, Sutskever, & Hinton, 2012;Ma, Zheng, Cheng, Zhang, & Han, 2019;Molina-Cabello, Luque-Baena, López-Rubio, & Thurnhofer-Hemsi, 2018;Rafiei & Adeli, 2017a;Wijnands, Nice, Thompson, Zhao, & Stevenson, 2019;Q. Zhao, Adeli, Honnorat, Leng, & Pohl, 2019).…”