Research Article: Do street-level scene perceptions affect housing prices in Chinese megacities? An analysis using open access datasets and deep learning

Date Published: May 30, 2019

Publisher: Public Library of Science

Author(s): Xiao Fu, Tianxia Jia, Xueqi Zhang, Shanlin Li, Yonglin Zhang, Shihe Fu.


Many studies have explored the relationship between housing prices and environmental characteristics using the hedonic price model (HPM). However, few studies have deeply examined the impact of scene perception near residential units on housing prices. This article used house purchasing records from and open access geolocation data (including massive street view pictures, point of interest (POI) data and road network data) and proposed a framework named “open-access-dataset-based hedonic price modeling (OADB-HPM)” for comprehensive analysis in Beijing and Shanghai, China. A state-of-the-art deep learning framework and massive Baidu street view panoramas were employed to visualize and quantify three major scene perception characteristics (greenery, sky and building view indexes, abbreviated GVI, SVI and BVI, respectively) at the street level. Then, the newly introduced scene perception characteristics were combined with other traditional characteristics in the HPM to calculate marginal prices, and the results for Beijing and Shanghai were explored and compared. The empirical results showed that the greenery and sky perceptual elements at the property level can significantly increase the housing price in Beijing (RMB 39,377 and 6011, respectively) and Shanghai (RMB 21,689 and 2763, respectively), indicating an objectively higher willingness by buyers to pay for houses that provide the ability to perceive natural elements in the surrounding environment. This study developed quantification tools to help decision makers and planners understand and analyze the interaction between residents and urban scene components.

Partial Text

Urban scenes can be considered a vital medium for human beings, and it is important to recognize and understand this aspect of cities [1] to ensure the spatial security of urban sustainability [2]. The spatial distributions of urban scenes exhibit variety, complexity and heterogeneity [3], and this variability is necessary for urban residents to survive and thrive [4]. The recognition, judgment, understanding and feelings associated with urban perception can impact people’s daily lives [5]. Therefore, urban perception is closely related to the amenities of the urban visual environment. For instance, residential parcels filled with rubbish and painted with graffiti can make pedestrians feel unsafe and further negatively affect residents’ willingness to live in the area [6,7]. Some studies have shown that the perceptual elements in the urban environment have a vital influence on people’s mental and physical health [8–10]. Visible greenery can reduce pedestrians’ stress and passive emotions [11] and can soothe hospital patients and shorten their recovery time [12]. Places such as urban green parks and open plazas can provide green landscapes and necessary spaces for residents and pedestrians to perform various physical activities [13], which can lower the risks of heart disease, diabetes and obesity [14] and improve people’s physical fitness [15]. Recent research has demonstrated that street-level visible greenery can significantly improve people’s likelihood to walk [16,17]. Although high-density high-rise buildings can enhance the oppressiveness experienced by pedestrians, green trees planted along roads can eliminate this negative effect [8,9]. In addition, recent studies have suggested that the perception of the thermal environment is connected to the openness of street canyons [18], which have an impact on both ground surface temperature and pedestrians’ walking comfort and physical health [18,19]. In summary, thoroughly studying and quantifying human-level urban environmental perceptions can help promote urban sustainability and public health.

This article used a state-of-the-art deep learning method to process massive street view panoramas to quantify the views of greenery, buildings and sky (GVI, BVI and SVI); then, these indicators were introduced into HPMs for the first time. This article mainly explored whether the newly included VIs affect the property-level housing prices in two typical Chinese megacities. We employed multisource open access datasets to quantify the locations and characteristics of residential neighborhoods. Taking the housing price data as the dependent variable, HPMs were built and included four major types of influencing characteristics: location, structure, neighborhood and scene perception characteristics. The empirical results showed that the street-level GVI and SVI values in Beijing and Shanghai both have a positive effect on housing price, and the corresponding marginal implicit prices were RMB 39,377 and 21,689, respectively, in Beijing and RMB 6011 and 2763, respectively, in Shanghai.