Learning Traffic Anomalies from Generative Models on Real-Time Observations

2025-02-03

AI Summary

Real-time anomaly detection in urban traffic is challenged by congestion, camera failures, visual occlusions, and adverse weather conditions. Method: This paper proposes an unsupervised end-to-end framework that pioneers the application of spatiotemporal generative adversarial networks (STGANs) to real-world city-scale video streams. It integrates graph neural networks (GNNs) with long short-term memory (LSTM) networks to model minute-level vehicle density time series, enabling fully automatic anomaly discovery without labeled anomaly data. Density-flow features are extracted via sliding-window-based real-time training and inference. Contribution/Results: Evaluated across 42 intersections in Gothenburg, Sweden, the framework achieves a false positive rate <3% on the November 2020 validation set, successfully detecting diverse real-world anomalies—including traffic signal failures, fog/snow occlusions, and sudden congestion—thereby significantly enhancing traffic situational awareness.

Technology Category

• Computer Vision: Generative Adversarial Networks (GANs) for Vision
• Planning, Routing, and Scheduling: Optimization of Spatio-temporal Systems
• Machine Learning: Time-Series/Data Streams

Application Category

• Graph Algorithms and Modeling for the Web: Graph neural networks and deep learning approaches for Web-related graphs
• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications
• Responsible Web: Human-perceived consequences of algorithmic deployment on the web

Abstract

Accurate detection of traffic anomalies is crucial for effective urban traffic management and congestion mitigation. We use the Spatiotemporal Generative Adversarial Network (STGAN) framework combining Graph Neural Networks and Long Short-Term Memory networks to capture complex spatial and temporal dependencies in traffic data. We apply STGAN to real-time, minute-by-minute observations from 42 traffic cameras across Gothenburg, Sweden, collected over several months in 2020. The images are processed to compute a flow metric representing vehicle density, which serves as input for the model. Training is conducted on data from April to November 2020, and validation is performed on a separate dataset from November 14 to 23, 2020. Our results demonstrate that the model effectively detects traffic anomalies with high precision and low false positive rates. The detected anomalies include camera signal interruptions, visual artifacts, and extreme weather conditions affecting traffic flow.

Problem

• Abnormal Traffic Detection
• Urban Traffic Management
• Congestion Reduction

Innovation

• STGAN
• Real-time Traffic Data
• Anomaly Detection