Image-Conditioned Graph Generation for Road Network Extraction

TL;DR

This paper introduces a novel deep autoregressive model called GGT for image-conditioned graph generation, specifically applied to extracting road networks from images, and provides a new dataset and evaluation metric for this task.

Contribution

The paper develops GGT, a new attention-based generative model for image-conditioned graph generation, and introduces the Toulouse Road Network dataset and StreetMover distance for evaluation.

Findings

GGT effectively generates road networks from semantic segmentation data.

The Toulouse Road Network dataset provides real-world data for benchmarking.

StreetMover distance offers a new way to evaluate road network generation quality.

Abstract

Deep generative models for graphs have shown great promise in the area of drug design, but have so far found little application beyond generating graph-structured molecules. In this work, we demonstrate a proof of concept for the challenging task of road network extraction from image data. This task can be framed as image-conditioned graph generation, for which we develop the Generative Graph Transformer (GGT), a deep autoregressive model that makes use of attention mechanisms for image conditioning and the recurrent generation of graphs. We benchmark GGT on the application of road network extraction from semantic segmentation data. For this, we introduce the Toulouse Road Network dataset, based on real-world publicly-available data. We further propose the StreetMover distance: a metric based on the Sinkhorn distance for effectively evaluating the quality of road network generation. The code and dataset are publicly available.