Hazard-Aware Traffic Scene Graph Generation

TL;DR

This paper introduces a new traffic scene graph generation task focusing on hazard relevance, proposing a framework that uses accident data and depth cues for improved hazard understanding in driving scenarios.

Contribution

It presents a novel hazard-aware traffic scene graph generation framework that incorporates traffic accident data and depth cues for better scene understanding.

Findings

Model effectively highlights prominent hazards with severity color-coding.

Framework demonstrates strong performance across 10 tasks and 5 evaluation perspectives.

Ablation studies confirm the importance of accident data and depth cues.

Abstract

Maintaining situational awareness in complex driving scenarios is challenging. It requires continuously prioritizing attention among extensive scene entities and understanding how prominent hazards might affect the ego vehicle. While existing studies excel at detecting specific semantic categories and visually salient regions, they lack the ability to assess safety-relevance. Meanwhile, the generic spatial predicates either for foreground objects only or for all scene entities modeled by existing scene graphs are inadequate for driving scenarios. To bridge this gap, we introduce a novel task, Traffic Scene Graph Generation, which captures traffic-specific relations between prominent hazards and the ego vehicle. We propose a novel framework that explicitly uses traffic accident data and depth cues to supplement visual features and semantic information for reasoning. The output traffic scene graphs provide intuitive guidelines that stress prominent hazards by color-coding their severity and notating their effect mechanism and relative location to the ego vehicle. We create relational annotations on Cityscapes dataset and evaluate our model on 10 tasks from 5 perspectives. The results in comparative experiments and ablation studies demonstrate our capacity in ego-centric reasoning for hazard-aware traffic scene understanding.