Building Footprint Generation Using Improved Generative Adversarial Networks

TL;DR

This paper introduces an improved conditional GAN approach with Wasserstein distance and gradient penalty for automatic building footprint generation from satellite images, significantly enhancing quality and reducing hyperparameter tuning.

Contribution

The work presents a novel GAN-based method that outperforms existing models in building footprint extraction and simplifies the tuning process.

Findings

Significant improvement over existing GAN and U-Net methods.

High-quality building footprint generation from satellite images.

Reduced need for hyperparameter tuning.

Abstract

Building footprint information is an essential ingredient for 3-D reconstruction of urban models. The automatic generation of building footprints from satellite images presents a considerable challenge due to the complexity of building shapes. In this work, we have proposed improved generative adversarial networks (GANs) for the automatic generation of building footprints from satellite images. We used a conditional GAN with a cost function derived from the Wasserstein distance and added a gradient penalty term. The achieved results indicated that the proposed method can significantly improve the quality of building footprint generation compared to conditional generative adversarial networks, the U-Net, and other networks. In addition, our method nearly removes all hyperparameters tuning.