RNGDet++: Road Network Graph Detection by Transformer with Instance Segmentation and Multi-scale Features Enhancement

Abstract

The graph structure of road networks is critical for downstream tasks of autonomous driving systems, such as global planning and navigation. In the past, the road network graph is usually manually annotated by human experts, which is time-consuming and labor-intensive. To obtain the road network graph with better effectiveness and efficiency, automatic approaches for road network graph detection are required. Previous works either post-process semantic segmentation maps or propose graph-based algorithms to directly predict the road network graph. However, previous works suffer from hard-coded processing algorithms and inferior final performance. To enhance the previous SOTA (State-of-the-Art) approach RNGDet, we add an instance segmentation head to better supervise the model training, and enable the model to leverage multi-scale features of the backbone network. Since the new proposed approach is improved from RNGDet, it is named RNGDet++. All approaches are evaluated on two large publicly available datasets. RNGDet++ outperforms baseline models on almost all metrics scores. It improves the topology correctness APLS (Average Path Length Similarity) by around 3%.

News

Mar/1/2023: Paper accepted by RA-L.

Dec/23/2022: We add experiments on the SpaceNet dataset. Please check the github repo. The manuscript and the supplementary document are also updated.

Paper

pdf

arxiv access

Supplymentary document

More details of our implementation and extra visualization are provided in our supplementary document.

Implementation Code

Inference code is release at our github page.