Patched Line Segment Learning for Vector Road Mapping

TL;DR

This paper introduces a novel patch-based line segment method for vector road mapping from satellite images, capturing road geometry efficiently and achieving state-of-the-art results with minimal training resources.

Contribution

The paper proposes a line segment representation for road graphs that simplifies construction and improves efficiency, outperforming existing methods in accuracy and training cost.

Findings

Achieves state-of-the-art performance on benchmark datasets.

Reduces training time by 32-fold compared to previous methods.

Effectively captures road geometry using patch-based line segments.

Abstract

This paper presents a novel approach to computing vector road maps from satellite remotely sensed images, building upon a well-defined Patched Line Segment (PaLiS) representation for road graphs that holds geometric significance. Unlike prevailing methods that derive road vector representations from satellite images using binary masks or keypoints, our method employs line segments. These segments not only convey road locations but also capture their orientations, making them a robust choice for representation. More precisely, given an input image, we divide it into non-overlapping patches and predict a suitable line segment within each patch. This strategy enables us to capture spatial and structural cues from these patch-based line segments, simplifying the process of constructing the road network graph without the necessity of additional neural networks for connectivity. In our experiments, we demonstrate how an effective representation of a road graph significantly enhances the performance of vector road mapping on established benchmarks, without requiring extensive modifications to the neural network architecture. Furthermore, our method achieves state-of-the-art performance with just 6 GPU hours of training, leading to a substantial 32-fold reduction in training costs in terms of GPU hours.