HiLO: High-Level Object Fusion for Autonomous Driving using Transformers

TL;DR

This paper introduces HiLO, a transformer-based high-level object fusion method for autonomous driving that improves perception accuracy while maintaining low computational complexity, validated on large-scale real-world data.

Contribution

The paper proposes a novel transformer-based high-level object fusion approach called HiLO, enhancing robustness and efficiency over traditional methods like Kalman filters.

Findings

25.9 percentage points improvement in F1 score

6.1 percentage points increase in mean IoU

Effective cross-domain generalization between urban and highway scenarios

Abstract

The fusion of sensor data is essential for a robust perception of the environment in autonomous driving. Learning-based fusion approaches mainly use feature-level fusion to achieve high performance, but their complexity and hardware requirements limit their applicability in near-production vehicles. High-level fusion methods offer robustness with lower computational requirements. Traditional methods, such as the Kalman filter, dominate this area. This paper modifies the Adapted Kalman Filter (AKF) and proposes a novel transformer-based high-level object fusion method called HiLO. Experimental results demonstrate improvements of $25.9$ percentage points in $\textrm{F}_1$ score and $6.1$ percentage points in mean IoU. Evaluation on a new large-scale real-world dataset demonstrates the effectiveness of the proposed approaches. Their generalizability is further validated by cross-domain evaluation between urban and highway scenarios. Code, data, and models are available at https://github.com/rst-tu-dortmund/HiLO .