VAGNet: Vision-based Accident Anticipation with Global Features

TL;DR

VAGNet is a deep neural network that predicts traffic accidents from dashcam videos using global scene features, achieving higher accuracy and efficiency than existing methods.

Contribution

It introduces a novel accident anticipation model leveraging global scene features with transformer and graph modules, eliminating the need for object-level feature extraction.

Findings

Outperforms existing methods in average precision and mean time-to-accident.

Operates in real-time with higher computational efficiency.

Validated on four benchmark datasets with superior results.

Abstract

Traffic accidents are a leading cause of fatalities and injuries across the globe. Therefore, the ability to anticipate hazardous situations in advance is essential. Automated accident anticipation enables timely intervention through driver alerts and collision avoidance maneuvers, forming a key component of advanced driver assistance systems. In autonomous driving, such predictive capabilities support proactive safety behaviors, such as initiating defensive driving and human takeover when required. Using dashcam video as input offers a cost-effective solution, but it is challenging due to the complexity of real-world driving scenes. Accident anticipation systems need to operate in real-time. However, current methods involve extracting features from each detected object, which is computationally intensive. We propose VAGNet, a deep neural network that learns to predict accidents from dash-cam video using global features of traffic scenes without requiring explicit object-level features. The network consists of transformer and graph modules, and we use the vision foundation model VideoMAE-V2 for global feature extraction. Experiments on four benchmark datasets (DAD, DoTA, DADA, and Nexar) show that our method anticipates accidents with higher average precision and mean time-to-accident while being computationally more efficient compared to existing methods.