Skip to Main content Skip to Navigation
Journal articles

3D Geological Image Synthesis From 2D Examples Using Generative Adversarial Networks

Guillaume Coiffier 1, * Philippe Renard 2 Sylvain Lefebvre 3
* Corresponding author
3 MFX - Matter from Graphics
LORIA - ALGO - Department of Algorithms, Computation, Image and Geometry, Inria Nancy - Grand Est
Abstract : Generative Adversarial Networks (GAN) are becoming an alternative to Multiple-point Statistics (MPS) techniques to generate stochastic fields from training images. But a difficulty for all the training image based techniques (including GAN and MPS) is to generate 3D fields when only 2D training data sets are available. In this paper, we introduce a novel approach called Dimension Augmenter GAN (DiAGAN) enabling GANs to generate 3D fields from 2D examples. The method is simple to implement and is based on the introduction of a random cut sampling step between the generator and the discriminator of a standard GAN. Numerical experiments show that the proposed approach provides an efficient solution to this long lasting problem.
Document type :
Journal articles
Complete list of metadata
Contributor : Sylvain Lefebvre Connect in order to contact the contributor
Submitted on : Friday, November 6, 2020 - 11:57:41 AM
Last modification on : Wednesday, November 3, 2021 - 7:56:50 AM

Links full text




Guillaume Coiffier, Philippe Renard, Sylvain Lefebvre. 3D Geological Image Synthesis From 2D Examples Using Generative Adversarial Networks. Frontiers in Water, Frontiers, 2020, 2, ⟨10.3389/frwa.2020.560598⟩. ⟨hal-02992194⟩



Record views