Implications of Artificial Intelligence for Assessing the Built Environment

The exponential rate of global urban growth makes evaluating numerous socio-spatial attributes of our everyday environment an increasingly daunting task. Consequently, many modern urban studies have taken advantage of artificial intelligence (AI), specifically machine learning and computer vision techniques, as preferred techniques for image-based assessment of the built environment. This article investigates the critical role of AI in urban studies, associated urban data collection methods, and preferred machine learning algorithms. Several implemented computer vision models are assessed for their accuracy, study area, and the urban attributes they evaluate. These attributes include neighborhood perception, child-friendliness, walkability, safety and security, aesthetics, urban forestry and greenery, land use, and transportation. Thereafter, the deficiencies and potential of the future applications of AI for assessing the built environment are discussed within the current context, which lacks comprehensive assessment models for evaluating critical aspects of urban life.