English

Driving Behavior Explanation with Multi-level Fusion

Machine Learning 2021-11-24 v1 Artificial Intelligence Computer Vision and Pattern Recognition Robotics

Abstract

In this era of active development of autonomous vehicles, it becomes crucial to provide driving systems with the capacity to explain their decisions. In this work, we focus on generating high-level driving explanations as the vehicle drives. We present BEEF, for BEhavior Explanation with Fusion, a deep architecture which explains the behavior of a trajectory prediction model. Supervised by annotations of human driving decisions justifications, BEEF learns to fuse features from multiple levels. Leveraging recent advances in the multi-modal fusion literature, BEEF is carefully designed to model the correlations between high-level decisions features and mid-level perceptual features. The flexibility and efficiency of our approach are validated with extensive experiments on the HDD and BDD-X datasets.

Keywords

Cite

@article{arxiv.2012.04983,
  title  = {Driving Behavior Explanation with Multi-level Fusion},
  author = {Hédi Ben-Younes and Éloi Zablocki and Patrick Pérez and Matthieu Cord},
  journal= {arXiv preprint arXiv:2012.04983},
  year   = {2021}
}

Comments

Accepted at NeurIPS Workshop ML4AD 2020

R2 v1 2026-06-23T20:50:29.984Z