Physics-Informed Neural Networks for Real- Time Crowd Density Prediction
Zhe Li, Xiangchuan Guo · 2024
With rapid urbanization and the increasing frequency of large-scale events, crowd safety management has become a critical concern. This paper proposes a novel approach based on Physics-Informed Neural Networks (PINNs) for real-time crowd density prediction. By incorporating fluid dynamics principles and domain knowledge into the framework, our model effectively captures the complex spatio-temporal dynamics of crowd motion and enables accurate density prediction in real-world scenarios. Experiments on a large-scale real-world dataset demonstrate the superiority of our approach over baselines in terms of both accuracy and efficiency.