Synapse
⌘+K
Synapse
PulseExploreClubsResearchersJournals
Instagram
HomeClubsExplore
November 1, 2025SPE Journal

High-Fidelity Seismic Image Generation From Limited Data Sets using Generative Adversarial Networks

View Full Paper
Ask AI
Bookmark
Share

Authors

ASA.I. SaidHRHughes RgMTM. Tyagi

Discussion

Loading...

Member takes

Overview

Analysis demonstrates enhanced seismic image quality in the North Sea, highlighting generative adversarial networks for resource management.

Key Points

  • Generative adversarial networks achieved high-fidelity seismic image generation, effectively overcoming data limitations.
  • Results indicated significant improvements, with FID scores down to 4.110, showcasing model capabilities in seismic imagery.
  • Assessment used a deep learning framework based on StyleGAN2, integrating conditional GANs and data augmentation for improved performance.
  • Findings suggest models can support various seismic applications, potentially transforming geophysical exploration and resource management.

Cite This Study

Said et al. (2025) studied this question.

synapsesocial.com/papers/6925437fc0ce034ddc358de8https://doi.org/10.2118/231402-pa
View Full Paper
Ask AI
Bookmark
Share

Also Consider

Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context:

  1. 1Multi-Condition Seismic Data Denoising using Feature-Expanded Gradient Penalty Generative Adversarial Network2025
  2. 2Cross‐Dimensional Generative Adversarial Networks (CDGAN) Geomodeling: Bridging 2D Geological Figures and 3D Reservoir Modeling2026
  3. 3Adversarial Augmented Fields for Efficient Geophysical Analysis2024
  4. 4Generative Adversarial Network for Modeling of CO2 Plume Evolution in Geological Carbon Storage Systems2025 · 2 citations
  5. 5Physics-informed W-Net GAN for the direct stochastic inversion of fullstack seismic data into facies models2024 · 5 citations