Advancing Deep Learning Predictions with Facies-Specific Porosity Realizations

Summary The aim of the study was to enhance the predictive capabilities of deep learning algorithms, specifically for 3D porosity prediction from seismic data. To achieve this, we generated geologically plausible porosity models to expand the initial training dataset. The resulting synthetic pseudo-wells improved the generalization of the deep learning models by offering a diverse and extensive training dataset that encompassed various subsurface scenarios, leading to more accurate porosity predictions.