Prior Based Online Lane Graph Extraction from Single Onboard Camera Image

TL;DR

This paper presents a novel online lane graph extraction method from a single onboard camera that leverages prior information via a transformer-based autoencoder to improve accuracy, tested on NuScenes and Argoverse datasets.

Contribution

It introduces a prior-based approach using a transformer autoencoder to enhance online lane graph estimation from monocular images.

Findings

Significant performance improvement over state-of-the-art methods.

Effective use of prior information to refine lane graph estimation.

Validated on two benchmark datasets, NuScenes and Argoverse.

Abstract

The local road network information is essential for autonomous navigation. This information is commonly obtained from offline HD-Maps in terms of lane graphs. However, the local road network at a given moment can be drastically different than the one given in the offline maps; due to construction works, accidents etc. Moreover, the autonomous vehicle might be at a location not covered in the offline HD-Map. Thus, online estimation of the lane graph is crucial for widespread and reliable autonomous navigation. In this work, we tackle online Bird's-Eye-View lane graph extraction from a single onboard camera image. We propose to use prior information to increase quality of the estimations. The prior is extracted from the dataset through a transformer based Wasserstein Autoencoder. The autoencoder is then used to enhance the initial lane graph estimates. This is done through optimization of the latent space vector. The optimization encourages the lane graph estimation to be logical by discouraging it to diverge from the prior distribution. We test the method on two benchmark datasets, NuScenes and Argoverse. The results show that the proposed method significantly improves the performance compared to state-of-the-art methods.