Fast and Scalable Earth Texture Synthesis using Spatially Assembled Generative Adversarial Neural Networks

TL;DR

This paper introduces SAGANs, a novel GAN-based method that efficiently synthesizes large-scale earth textures like rocks, maintaining structural properties and significantly reducing computational costs compared to traditional GANs.

Contribution

The paper proposes a spatially assembled GAN framework that enables scalable, large-size earth texture synthesis with improved efficiency and structural fidelity.

Findings

SAGANs can generate arbitrarily large geological textures.

The method maintains topological and structural properties of training images.

Computational efficiency is significantly improved over standard GANs.

Abstract

The earth texture with complex morphological geometry and compositions such as shale and carbonate rocks, is typically characterized with sparse field samples because of an expensive and time-consuming characterization process. Accordingly, generating arbitrary large size of the geological texture with similar topological structures at a low computation cost has become one of the key tasks for realistic geomaterial reconstruction. Recently, generative adversarial neural networks (GANs) have demonstrated a potential of synthesizing input textural images and creating equiprobable geomaterial images. However, the texture synthesis with the GANs framework is often limited by the computational cost and scalability of the output texture size. In this study, we proposed a spatially assembled GANs (SAGANs) that can generate output images of an arbitrary large size regardless of the size of training images with computational efficiency. The performance of the SAGANs was evaluated with two and three dimensional (2D and 3D) rock image samples widely used in geostatistical reconstruction of the earth texture. We demonstrate SAGANs can generate the arbitrary large size of statistical realizations with connectivity and structural properties similar to training images, and also can generate a variety of realizations even on a single training image. In addition, the computational time was significantly improved compared to standard GANs frameworks.