Progressively Growing Generative Adversarial Networks for High\n Resolution Semantic Segmentation of Satellite Images

2019/02/12 by Edward Collier, Kate Duffy, Collier, Edward +19
Computer Science · Environmental Science · #Advanced Neural Network Applications #Computer Vision and Pattern Recognition (cs.CV) #FOS: Computer and information sciences #Generative Adversarial Networks and Image Synthesis #Remote Sensing and LiDAR Applications #Video Surveillance and Tracking Methods

paper · pdf · doi:10.48550/arxiv.1902.04604

openalex publication_date 2019/02/12 · openalex created_date 2022/07/29 · openalex updated_date 2026/07/28

Abstract

Machine learning has proven to be useful in classification and segmentation\nof images. In this paper, we evaluate a training methodology for pixel-wise\nsegmentation on high resolution satellite images using progressive growing of\ngenerative adversarial networks. We apply our model to segmenting building\nrooftops and compare these results to conventional methods for rooftop\nsegmentation. We present our findings using the SpaceNet version 2 dataset.\nProgressive GAN training achieved a test accuracy of 93% compared to 89% for\ntraditional GAN training.\n

Citations

• Overcoming Missing and Incomplete Modalities with Generative Adversarial\n Networks for Building Footprint Segmentation

Related

• Using GANs to Augment Data for Cloud Image Segmentation Task
• Semantic Segmentation of Medium-Resolution Satellite Imagery using Conditional Generative Adversarial Networks
• Super-Resolution for Overhead Imagery Using DenseNets and Adversarial\n Learning
• Application of Generative Adversarial Network (GAN) for Synthetic Training Data Creation to improve performance of ANN Classifier for extracting Built-Up pixels from Landsat Satellite Imagery
• Can Location Embeddings Enhance Super-Resolution of Satellite Imagery?