DRIVE: Deep Reinforced Accident Anticipation with Visual Explanation
This addresses the problem of improving safety in self-driving systems by enabling early and explainable accident prediction, though it is incremental as it builds on existing anticipation approaches.
The paper tackles traffic accident anticipation from dashcam videos by proposing DRIVE, a method that uses deep reinforcement learning with visual attention mechanisms to provide explanations for its decisions, achieving state-of-the-art performance on multiple real-world datasets.
Traffic accident anticipation aims to accurately and promptly predict the occurrence of a future accident from dashcam videos, which is vital for a safety-guaranteed self-driving system. To encourage an early and accurate decision, existing approaches typically focus on capturing the cues of spatial and temporal context before a future accident occurs. However, their decision-making lacks visual explanation and ignores the dynamic interaction with the environment. In this paper, we propose Deep ReInforced accident anticipation with Visual Explanation, named DRIVE. The method simulates both the bottom-up and top-down visual attention mechanism in a dashcam observation environment so that the decision from the proposed stochastic multi-task agent can be visually explained by attentive regions. Moreover, the proposed dense anticipation reward and sparse fixation reward are effective in training the DRIVE model with our improved reinforcement learning algorithm. Experimental results show that the DRIVE model achieves state-of-the-art performance on multiple real-world traffic accident datasets. Code and pre-trained model are available at \url{https://www.rit.edu/actionlab/drive}.