Search

The acceleration of urban imagery data analysis, driven by computer vision (CV), has created noteworthy opportunities for urban studies and planning. Data on street environments with high granularity derived from geo-tagged street views allow urban researchers to obtain geospatial data on greenery, pavement materials, and dimensions, building facades, urban furniture, lighting, vehicle presence, etc. However, how such attributes have been classified and used to address urban studies, planning, or mobility questions remains relatively poorly understood among non-technical researchers. Targeting urban planning and design researchers who do not have a background in CV, this paper reviews planning-relevant attributes that CV approaches of urban streetscapes have delivered to date and examines some of their research applications. We present a systematic analysis of 146 papers scrutinizing 104 street attributes in four groups. By exploring a subcollection of 24 papers, we discuss the effectiveness of those attributes being incorporated into current quantitative urban studies. This study's primary contribution lies in providing a comprehensive summary of CV-driven street attributes, their applications, and the algorithms used, serving as a valuable resource for future urban research. Additionally, we identify key challenges in this field, such as unclear definitions of attributes, a disproportionate emphasis on selecting models and features, and the absence of standardized measurement and definition methods. Furthermore, we offer recommendations for future research directions in this area. • We review 104 street attributes collected with computer vision in urban studies • We summarize 23 chosen attributes' correlations with known urban design measures • We provide a table of CV-driven street attributes with model details, references, and impacts • We highlight challenges in attribute definitions, research focus, and impact clarity • We provide recommendations for standardizing data and future research [ABSTRACT FROM AUTHOR]

As municipalities are setting ambitious targets to increase non-motorized transportation mode shares, analytic frameworks for examining how such targets will be met become increasingly important. In this article, we update the Urban Network Analysis pedestrian modeling framework designed to link land use and urban development changes with pedestrian mobility outcomes by introducing new methods for capturing the effects of street properties on pedestrian travel demand and adjusting pedestrian trip generation rates with k-nearest accessibility scores based on destination availability in each context. This framework was used in conjunction with a participatory design process in Beirut, Lebanon to estimate pedestrian mobility impacts of three urban design scenarios. We illustrate how each scenario affects pedestrian trip generation across various trip types and trip distribution on individual street segments. Our approach demonstrates how urban design interventions–both land use changes and street quality improvements–can influence pedestrian travel demand. Estimates of these changes can provide planners and policymakers with valuable benefit-cost analyses of public space improvements, and a framework for understanding how site-specific planning and development decisions can impact progress towards (or away from) non-motorized mobility goals. • We propose a pedestrian impact assessment framework to estimate how street or building changes affect pedestrian mobility. • We introduce "perceived" route lengths to account for street attribute effects on pedestrian travel demand. • We introduce the concept of of k-nearest destination accessibility, which enables elasticities in pedestrian trip generation. • The framework is used to estimate urban design impacts on trip distribution, and for the first time, also trip generation. • Using Beirut, Lebanon as example, the framework can guide development options to decarbonize urban mobility. [ABSTRACT FROM AUTHOR]

This article analyzes location patterns of retail and food establishments, whose presence constitutes an important aspect of livable and walkable neighborhoods. Using approximately fourteen thousand buildings on the street network of Cambridge and Somerville, Massachusetts, as units of analysis, this article tests five hypotheses about retail locations found in previous literature and estimates the impacts of different location characteristics for all retail and food-service establishment as a group and for different store categories individually. The results describe how specific location attributes impact the probabilities of finding retailers, which can inform the planning of commercial activity clusters. [ABSTRACT FROM PUBLISHER]

Advances in computational urbanism have stimulated the rise of generative and parametric approaches to urban design. Yet, most generative and parametric approaches focus on physical characteristics, such as a neighborhoods walkability, energy efficiency, and urban form. Here, we study the colocation patterns of more than one million amenities in 47 U.S. cities to model the amenity mix of neighborhoods, and to identify the amenities that are over- or under-supplied in a neighborhood. We build this model by combining a clustering algorithm, designed to identify amenity-dense neighborhoods, and a network, connecting amenities that are likely to collocate. Our findings extend generative and parametric urban design approaches to the amenity mix of neighborhoods, by leveraging the idea of relatedness from the economic geography literature, to evaluate and optimize a neighborhood's amenity mix. [ABSTRACT FROM AUTHOR]