Urban Safety

Urban safety research focuses on improving the accuracy and efficiency of assessing and predicting safety levels in urban environments, primarily to enhance resource allocation and proactive crime prevention. Current approaches leverage machine learning, particularly models like Random Forests and neural networks, analyzing diverse data sources including police reports, street-level imagery, and crowdsourced safety perceptions. These methods aim to create more objective and data-driven assessments of urban safety, surpassing traditional subjective methods, and informing urban planning and policy decisions for improved public safety outcomes. The integration of multimodal large language models shows promise in automating safety perception assessments, offering a significant advancement in scalability and efficiency.

Papers

September 17, 2024

Machine Learning for Public Good: Predicting Urban Crime Patterns to Enhance Community Safety
Sia Gupta, Simeon Sayer
Machine Learning Crime Prediction Crime Pattern Urban Safety

July 29, 2024

Revolutionizing Urban Safety Perception Assessments: Integrating Multimodal Large Language Models with Street View Images
Jiaxin Zhang, Yunqin Li, Tomohiro Fukuda, Bowen Wang
Multimodal Large Language Model Street View Urban Safety

April 15, 2024

Eyes on the Streets: Leveraging Street-Level Imaging to Model Urban Crime Dynamics
Zhixuan Qi, Huaiying Luo, Chen Chi
Character Eye Urban Scene Urban Planning Crime Pattern Urban STREET Urban Safety

August 4, 2023

SURE-Val: Safe Urban Relevance Extension and Validation
Kai Storms, Ken Mori, Steven Peters
External Validation Relevance Modeling Perception Datasets Relevance Judgment Yes Yes Perceptual Concept Urban Safety

November 19, 2022

Evaluating the Perceived Safety of Urban City via Maximum Entropy Deep Inverse Reinforcement Learning
Yaxuan Wang, Zhixin Zeng, Qijun Zhao
Reinforcement Learning Reward Function Inverse Reinforcement Learning Urban Environment Safety Climate Perception Urban Safety Urban Space Perception

Urban Safety

Papers

Machine Learning for Public Good: Predicting Urban Crime Patterns to Enhance Community Safety

Revolutionizing Urban Safety Perception Assessments: Integrating Multimodal Large Language Models with Street View Images

Eyes on the Streets: Leveraging Street-Level Imaging to Model Urban Crime Dynamics

SURE-Val: Safe Urban Relevance Extension and Validation

Evaluating the Perceived Safety of Urban City via Maximum Entropy Deep Inverse Reinforcement Learning