Real-time Accident Anticipation for Autonomous Driving Through Monocular Depth-Enhanced 3D Modeling
This work addresses the critical safety problem of accident anticipation for autonomous driving systems, representing a strong specific gain over existing methods.
The study tackled real-time traffic accident anticipation for autonomous driving by introducing AccNet, a framework that uses monocular depth cues for 3D modeling and a novel loss function, achieving superior predictive accuracy on benchmark datasets with metrics like Average Precision and mean Time-To-Accident.
The primary goal of traffic accident anticipation is to foresee potential accidents in real time using dashcam videos, a task that is pivotal for enhancing the safety and reliability of autonomous driving technologies. In this study, we introduce an innovative framework, AccNet, which significantly advances the prediction capabilities beyond the current state-of-the-art (SOTA) 2D-based methods by incorporating monocular depth cues for sophisticated 3D scene modeling. Addressing the prevalent challenge of skewed data distribution in traffic accident datasets, we propose the Binary Adaptive Loss for Early Anticipation (BA-LEA). This novel loss function, together with a multi-task learning strategy, shifts the focus of the predictive model towards the critical moments preceding an accident. {We rigorously evaluate the performance of our framework on three benchmark datasets--Dashcam Accident Dataset (DAD), Car Crash Dataset (CCD), and AnAn Accident Detection (A3D), and DADA-2000 Dataset--demonstrating its superior predictive accuracy through key metrics such as Average Precision (AP) and mean Time-To-Accident (mTTA).