A platform for research: civil engineering, architecture and urbanism
A Visual Preference Analysis of Building Façades in Street View Images: A Case Study of Nanshan District, Shenzhen
Human visual perception and visual preferences affect how people perceive and experience the urban environment. Constrained by their methodologies and data availability, the previous studies have struggled to precisely quantify the subjects and objects in urban perception. This research employs visual saliency detection and fractal dimension measurement to quantify the characteristics of urban building façades and visual perception. Through this study, we determine the differences between human visual perception and computer vision. The study shows that (1) human vision exhibits selective preferences, processing building façades with varying complexity levels to form a critical complexity threshold, which is a feature that is not captured by computer vision processing; (2) taking Nanshan District, Shenzhen, as a case study, the value of this threshold is 1.2383; (3) building façades with complexity greater than this threshold are seen as “complex” building façades and vice versa; (4) when perceiving “simple” buildings, human eyes tend to focus on their more complex local areas, whereas for “complex” buildings, they pay more attention to simpler parts. This study provides a reference for conducting quantitative research on urban perception and visual perception.
A Visual Preference Analysis of Building Façades in Street View Images: A Case Study of Nanshan District, Shenzhen
Human visual perception and visual preferences affect how people perceive and experience the urban environment. Constrained by their methodologies and data availability, the previous studies have struggled to precisely quantify the subjects and objects in urban perception. This research employs visual saliency detection and fractal dimension measurement to quantify the characteristics of urban building façades and visual perception. Through this study, we determine the differences between human visual perception and computer vision. The study shows that (1) human vision exhibits selective preferences, processing building façades with varying complexity levels to form a critical complexity threshold, which is a feature that is not captured by computer vision processing; (2) taking Nanshan District, Shenzhen, as a case study, the value of this threshold is 1.2383; (3) building façades with complexity greater than this threshold are seen as “complex” building façades and vice versa; (4) when perceiving “simple” buildings, human eyes tend to focus on their more complex local areas, whereas for “complex” buildings, they pay more attention to simpler parts. This study provides a reference for conducting quantitative research on urban perception and visual perception.
A Visual Preference Analysis of Building Façades in Street View Images: A Case Study of Nanshan District, Shenzhen
Chia-Chen Lee (author) / Yuxiao Wang (author) / Chenggao Tang (author) / Xiang Li (author) / Jie Yin (author)
2025
Article (Journal)
Electronic Resource
Unknown
Metadata by DOAJ is licensed under CC BY-SA 1.0