Style transfer for generation of realistically textured subsurface models
KAUST DepartmentEarth Science and Engineering Program
Physical Science and Engineering (PSE) Division
Extreme Computing Research Center
Computer, Electrical and Mathematical Sciences and Engineering (CEMSE) Division
Permanent link to this recordhttp://hdl.handle.net/10754/667620
MetadataShow full item record
AbstractTraining datasets consisting of numerous pairs of subsurface models and target variables are essential for building machine learning solutions for geophysical applications. We apply an iterative style transfer approach from image processing to produce realistically textured subsurface models based on synthetic prior models. The key idea of style transfer is that content and texture representations within a convolutional neural network are, to some extent, separable. Thus, a style from one image can be transferred to match the content from another image. We demonstrate examples where realistically random models are stylized to mimic texture patterns from Marmousi II and a section from the BP 2004 benchmark velocity models.
CitationOvcharenko, O., Kazei, V., Peter, D., & Alkhalifah, T. (2019). Style transfer for generation of realistically textured subsurface models. SEG Technical Program Expanded Abstracts 2019. doi:10.1190/segam2019-3216349.1
SponsorsWe thank Kevin Zakka for his implementation of the Gatys et al. (2015) algorithm (https://github.com/kevinzakka/style-transfer). The research reported in this publication was supported by funding from King Abdullah University of Science and Technology (KAUST), Thuwal, 23955-6900, Saudi Arabia.
PublisherSociety of Exploration Geophysicists
Conference/Event nameSociety of Exploration Geophysicists International Exposition and Annual Meeting 2019, SEG 2019