Abstract
In the last 40 years, Urban perception has become an important research area covering several fields, such as criminology, psychology, urban planning, Broken windows theory. It aims to analyze and interpret the behavior of the perception in cities. Urban perception focuses on understanding urban environments based on the characteristics of the city. With the rapidly increasing data availability and highly scalable data collection methods powered by modern web services, new techniques from other domains enabled the exploration of solutions to estimate urban perception (i.e., quantify urban perception autonomously). This work presents a methodology to explore the urban perception analysis task. The work relies on the benchmark dataset, Place Pulse. This dataset is used to perform our classification tasks concerning the category of safety in urban perception problems.
Materials
BibTeX
@inproceedings{2021-UrbanPerceptionQuantify,
title = {Quantifying Urban Safety Perception on Street View Images},
author = {Felipe Moreno-Vera AND Bahram Lavi AND Jorge Poco},
booktitle = {International Conference on Web Intelligence and Intelligent Agent Technology (WI-IAT)},
year = {2021},
url = {http://www.visualdslab.com/papers/UrbanPerceptionQuantify},
}