STAGNet: A Spatio-Temporal Graph and LSTM Framework for Accident Anticipation

TL;DR

This paper introduces STAGNet, a novel framework combining spatio-temporal graph features and LSTM networks to predict accidents from dash-cam videos, improving safety predictions in autonomous driving systems.

Contribution

The paper presents a new spatio-temporal graph and LSTM-based framework that outperforms existing methods in accident anticipation from dash-cam videos.

Findings

Achieves higher average precision in accident prediction

Improves mean time-to-accident scores

Performs well across multiple datasets

Abstract

Accident prediction and timely preventive actions improve road safety by reducing the risk of injury to road users and minimizing property damage. Hence, they are critical components of advanced driver assistance systems (ADAS) and autonomous vehicles. While many existing systems depend on multiple sensors such as LiDAR, radar, and GPS, relying solely on dash-cam videos presents a more challenging, yet more cost-effective and easily deployable solution. In this work, we incorporate improved spatio-temporal features and aggregate them through a recurrent network to enhance state-of-the-art graph neural networks for predicting accidents from dash-cam videos. Experiments using three publicly available datasets (DAD, DoTA and DADA) show that our proposed STAGNet model achieves higher average precision and mean time-to-accident scores than previous methods, both when cross-validated on a given dataset and when trained and tested on different datasets.