AI RESEARCH
RiskProp: Collision-Anchored Self-Supervised Risk Propagation for Early Accident Anticipation
arXiv CS.CV
•
ArXi:2603.27165v1 Announce Type: new Accident anticipation aims to predict impending collisions from dashcam videos and trigger early alerts. Existing methods rely on binary supervision with manually annotated "anomaly onset" frames, which are subjective and inconsistent, leading to inaccurate risk estimation. In contrast, we propose RiskProp, a novel collision-anchored self-supervised risk propagation paradigm for early accident anticipation, which removes the need for anomaly onset annotations and leverages only the reliably annotated collision frame.