Multi-Modal Fusion Transformer
for End-to-End Autonomous Driving

CVPR 2021

  • Max Planck Institute for Intelligent Systems
  • University of Tübingen
Abstract

How should representations from complementary sensors be integrated for autonomous driving? Geometry-based sensor fusion has shown great promise for perception tasks such as object detection and motion forecasting. However, for the actual driving task, the global context of the 3D scene is key, e.g. a change in traffic light state can affect the behavior of a vehicle geometrically distant from that traffic light. Geometry alone may therefore be insufficient for effectively fusing representations in end-to-end driving models. In this work, we demonstrate that imitation learning policies based on existing sensor fusion methods under-perform in the presence of a high density of dynamic agents and complex scenarios, which require global contextual reasoning, such as handling traffic oncoming from multiple directions at uncontrolled intersections. Therefore, we propose TransFuser, a novel Multi-Modal Fusion Transformer, to integrate image and LiDAR representations using attention. We experimentally validate the efficacy of our approach in urban settings involving complex scenarios using the CARLA urban driving simulator. Our approach achieves state-of-the-art driving performance while reducing collisions by 76% compared to geometry-based fusion.

Video
Generalization to New Town
Generalization to New Weather
Infraction Analysis
Attention Map Visualizations

For the yellow source token, we show the top-5 attended tokens in green and highlight the presence of vehicles in the LiDAR point cloud in red. TransFuser focuses on objects of interest (vehicles and traffic lights) at intersections, albeit at a slightly different location.

Citation

@inproceedings{Prakash2021CVPR,
author = {Prakash, Aditya and Chitta, Kashyap and Geiger, Andreas},
title = {Multi-Modal Fusion Transformer for End-to-End Autonomous Driving},
booktitle = {IEEE Conference on Computer Vision and Pattern Recognition (CVPR)},
year = {2021}
}

Template from this website