using visual AI to identify urban uniqueness
We first assess the identity of each city by grouping daily streetscapes into eight clusters of significant visual elements, deriving their unique compositions.
We next assess the distinctiveness of each city through determining the prevalence of the clusters in each city compared to the others.
Combining the visual identity and distinctiveness, we uncover the uniqueness of each city.
Each city has its own unique character. While Kyoto stands out as the most unique, Fukuoka shares the most visual similarities with other Japanese cities.
In Japan Uniqueness we utilize street view imagery and a
landmark-free framework to analyze how different visual
patterns build up the unique characteristics of cities. Here we
focus on the six most visited Japanese cities and found that eight
representative visual clusters explain each city's visual identity
and relative distinctiveness.
Our research demonstrates how artificial intelligence applied to
visual data can reveal subtle differences in urban environments.
In the era of growing globalization, the cultivation of a city's
unique visual characteristics can help avoid the homogenization of
urban landscapes, and stimulate the development of placemaking
strategies.
Guo, S., Jang, KM., Duarte, F., Ratti, C. (2025). Urban Visual Uniqueness: A Landmark-Free Framework to Quantify City's Identity and Distinctiveness from Everyday Scenes . Computers, Environment and Urban Systems
The material on this website can be used freely in any publication provided that:
For more information,
senseable-contacts@mit.edu