Learning Lane Graphs from Aerial Imagery Using Transformers

TL;DR

This paper presents a transformer-based method for generating detailed lane graphs from aerial imagery, improving autonomous vehicle navigation in complex urban environments.

Contribution

It introduces a novel approach using Detection Transformers to generate successor lane graphs directly from aerial images, a capability not previously explored.

Findings

High accuracy in generating successor lane graphs

Effective on large-scale UrbanLaneGraph dataset

Potential to improve autonomous navigation in complex settings

Abstract

The robust and safe operation of automated vehicles underscores the critical need for detailed and accurate topological maps. At the heart of this requirement is the construction of lane graphs, which provide essential information on lane connectivity, vital for navigating complex urban environments autonomously. While transformer-based models have been effective in creating map topologies from vehicle-mounted sensor data, their potential for generating such graphs from aerial imagery remains untapped. This work introduces a novel approach to generating successor lane graphs from aerial imagery, utilizing the advanced capabilities of transformer models. We frame successor lane graphs as a collection of maximal length paths and predict them using a Detection Transformer (DETR) architecture. We demonstrate the efficacy of our method through extensive experiments on the diverse and large-scale UrbanLaneGraph dataset, illustrating its accuracy in generating successor lane graphs and highlighting its potential for enhancing autonomous vehicle navigation in complex environments.