Athens Graffiti 雅典涂鸦
This study explores the correlation between street grafiti and spatial cognitive indicators in the first district of Athens, and transform all those features into a travel route planning board game for more engaging and intuitive learning of Athens city.
Based on more than 30,000 street-view scenes collected from over 7,000 geographical locations, graffiti conditions were classified into four categories, while street-level spatial cognition was evaluated across six perceptual dimensions. The results were visualized through spatial mapping.
Further analysis using regression models, correlation tests, and other statistical methods reveals that graffiti presence is positively correlated with the perception of “liveliness and depression”, and negatively correlated with perceptions of “safety, cleanliness, wealth, and beauty”. These findings highlight the complex and ambivalent impact of graffiti on urban spatial cognition.
本项目探讨了雅典第一区中街头涂鸦与空间认知之间的关系，并进一步将研究结果转化为一款旅行路线规划桌游，通过趣味游戏互动帮助游客更好理解雅典城市街道。
研究基于来自 7,000 余个地理位置的 30,000 多张街景图像，将涂鸦状态划分为 四类，并从六个感知维度对街道层面的空间认知进行评估，最终通过地图映射的方式对结果进行可视化呈现。进一步的，我们采用了回归分析、相关性检验等统计方法，发现：涂鸦状态与“活跃—压抑”感知指标呈显著正相关，而与“安全、整洁、富裕与美观”等感知指标呈负相关，揭示了街头涂鸦对城市空间认知所产生的复杂且矛盾的影响。
Paper Accepted by ASCAAD 2024 Conference
X. Guan, Y. Gao, J. Cai, X. Xu, and H. Tu,
Athens-Polis: Exploring the Correlation between Urban Graffiti and Cognitive Indices in Athens Through Machine Learning.
ASCAAD 2024
Available:：https://papers.cumincad.org/cgi-bin/works/paper/ascaad2024_027
01| Idea
Educational Board Game
Modern artistic graffiti, intertwined with the city’s ancient and elegant character, has emerged as a new urban symbol. Athens, one of the most densely graffiti-covered cities in the world, has seen graffiti become an indispensable element of its urban landscape amid ongoing historical and modern transformations.
This project analyzes graffiti types and spatial perception scores across the city and visualizes them through a board game map.
Through interactive gameplay, players are challenged to plan a high-scoring graffiti route based on the sequence of cards drawn, enabling them to learn about the city’s spatial characteristics through decision-making and exploration. The game provides an intuitive and engaging framework for understanding urban space.
02| Method
Machine Learning
By training and evaluating image classification models, the study captures the city’s characteristics based on large-scale real-world data. Building on this foundation, we further examine the correlations between perception scores and different graffiti types, revealing how graffiti shapes urban spatial cognition.