Over the past five decades, China has witnessed rapid urbanization. Since the 1970s, Beijing’s population has grown 150%, its subway network expanded from a single line covering 10.7 kilometers to 27 lines spanning 836 kilometers, and its housing stock has increased by 95.3%. These significant urban changes have paralleled national economic and social transformations, notably after China opened its markets to international trade in 1978. This study examines the repercussions of these broad urban and socio-political changes on individual domestic spaces in Beijing. Analyzing floor plans of over 2000 apartments built between 1970 and 2020, we introduce a computer vision-aided methodology to quantitatively assess the distribution, intensity, and nature of domestic activities, employing isovist analysis while considering property size and age. This approach facilitates a nuanced understanding of changes in visual accessibility for specific room types, thereby shedding light on the evolution of social life as reflected in the spatial configurations of the pre- and post-housing privatization eras. Our findings show a trend in the isovist intensity of living rooms, dining rooms, and kitchens, indicative of shifts toward privacy and mixed-use spaces, shaped by household occupancy patterns. By harnessing numerous online home interior images, this study underscores the potential of integrating large-scale imagery data with computer vision technology to yield profound insights into architectural and domestic studies.
Beijing’s urbanization reflected in apartment layouts: a computer vision and isovist analysis (1970–2020)
Martina Mazzarello;
2025-01-01
Abstract
Over the past five decades, China has witnessed rapid urbanization. Since the 1970s, Beijing’s population has grown 150%, its subway network expanded from a single line covering 10.7 kilometers to 27 lines spanning 836 kilometers, and its housing stock has increased by 95.3%. These significant urban changes have paralleled national economic and social transformations, notably after China opened its markets to international trade in 1978. This study examines the repercussions of these broad urban and socio-political changes on individual domestic spaces in Beijing. Analyzing floor plans of over 2000 apartments built between 1970 and 2020, we introduce a computer vision-aided methodology to quantitatively assess the distribution, intensity, and nature of domestic activities, employing isovist analysis while considering property size and age. This approach facilitates a nuanced understanding of changes in visual accessibility for specific room types, thereby shedding light on the evolution of social life as reflected in the spatial configurations of the pre- and post-housing privatization eras. Our findings show a trend in the isovist intensity of living rooms, dining rooms, and kitchens, indicative of shifts toward privacy and mixed-use spaces, shaped by household occupancy patterns. By harnessing numerous online home interior images, this study underscores the potential of integrating large-scale imagery data with computer vision technology to yield profound insights into architectural and domestic studies.| File | Dimensione | Formato | |
|---|---|---|---|
|
20250320_Yao-etal_BeijingUrbanization_HSSC.pdf
accesso aperto
Dimensione
26.57 MB
Formato
Adobe PDF
|
26.57 MB | Adobe PDF | Visualizza/Apri |
I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.


