About cookies on this site Our websites require some cookies to function properly (required). In addition, other cookies may be used with your consent to analyze site usage, improve the user experience and for advertising. For more information, please review your options. By visiting our website, you agree to our processing of information as described in IBM’sprivacy statement. To provide a smooth navigation, your cookie preferences will be shared across the IBM web domains listed here.
Publication
EAGE 2019
Conference paper
Synthetic seismic images from simple sketches using deep generative networks
Abstract
Several works have investigated the use of sketches to facilitate the creation of models by providing a faster and more intuitive tool set. Approaches that are already consolidated in domains such as architecture have proved to be much more difficult when applied to 3D geological modelling, given the size of the solution space. Nevertheless, more specific sketch-based applications have been proposed in geosciences such as seismic horizon modelling, geological storytelling, 3D fluvial system modelling and a tentative generic 3D modelling tool for geology. In this work, we investigate the applicability of deep generative networks for a simpler task: synthesizing seismic sections. Following the rising interest of the geoscience community in generative models such as Generative Adversarial Networks (GANs), we train a conditional GAN to generate seismic sections from simple sketches. The results are promising and find application in many tasks such as sketch-based seismic image retrieval and the generation of training data for machine learning algorithms such as Convolutional Neural Networks (CNNs).