CG-Net: Conditional GIS-aware Network for Individual Building Segmentation in VHR SAR Images

TL;DR

This paper introduces CG-Net, a novel neural network that leverages GIS building footprints to improve individual building segmentation in VHR SAR images, demonstrating robustness and potential for 3D reconstruction.

Contribution

We propose CG-Net, a GIS-aware neural network that effectively integrates building footprint data for enhanced SAR image segmentation, including methods for handling data inaccuracies and generating large-scale datasets.

Findings

CG-Net improves segmentation accuracy across different backbones.

Using complete building footprints yields better results than sensor-visible segments.

CG-Net remains robust against GIS data positioning errors.

Abstract

Object retrieval and reconstruction from very high resolution (VHR) synthetic aperture radar (SAR) images are of great importance for urban SAR applications, yet highly challenging owing to the complexity of SAR data. This paper addresses the issue of individual building segmentation from a single VHR SAR image in large-scale urban areas. To achieve this, we introduce building footprints from GIS data as complementary information and propose a novel conditional GIS-aware network (CG-Net). The proposed model learns multi-level visual features and employs building footprints to normalize the features for predicting building masks in the SAR image. We validate our method using a high resolution spotlight TerraSAR-X image collected over Berlin. Experimental results show that the proposed CG-Net effectively brings improvements with variant backbones. We further compare two representations of building footprints, namely complete building footprints and sensor-visible footprint segments, for our task, and conclude that the use of the former leads to better segmentation results. Moreover, we investigate the impact of inaccurate GIS data on our CG-Net, and this study shows that CG-Net is robust against positioning errors in GIS data. In addition, we propose an approach of ground truth generation of buildings from an accurate digital elevation model (DEM), which can be used to generate large-scale SAR image datasets. The segmentation results can be applied to reconstruct 3D building models at level-of-detail (LoD) 1, which is demonstrated in our experiments.