Full-waveform variational inference with full common-image gathers and diffusion network

2025/04/06 by Huseyin Tuna Erdinc, Zeng, Yunlin, Erdinc, Huseyin Tuna +4 · 3 citations
Earth and Planetary Sciences · Engineering · #FOS: Physical sciences #Geophysics (physics.geo-ph) #Hydrocarbon exploration and reservoir analysis #Image Processing Techniques and Applications #Seismic Imaging and Inversion Techniques

paper · pdf · doi:10.48550/arxiv.2504.15289

openalex publication_date 2025/04/06 · openalex created_date 2025/10/10 · openalex updated_date 2026/07/28

Abstract

Accurate seismic imaging and velocity estimation are essential for subsurface characterization. Conventional inversion techniques, such as full-waveform inversion, remain computationally expensive and sensitive to initial velocity models. To address these challenges, we propose a simulation-based inference framework with conditional elucidated diffusion models for posterior velocity-model sampling. Our approach incorporates both horizontal and vertical subsurface offset common-image gathers to capture a broader range of reflector geometries, including gently dipping structures and steep dipping layers. Additionally, we introduce the background-velocity model as an input condition to enhance generalization across varying geological settings. We evaluate our method on the SEAM dataset, which features complex salt geometries, using a patch-based training approach. Experimental results demonstrate that adding the background-velocity model as an additional conditioning variable significantly enhances performance, improving SSIM from 0.717 to 0.733 and reducing RMSE from 0.381 km/s to 0.274 km/s. Furthermore, uncertainty quantification analysis shows that our proposed approach yields better-calibrated uncertainty estimates, reducing uncertainty calibration error from 6.68 km/s to 3.91 km/s. These results show robust amortized seismic inversion with uncertainty quantification.

Citations

• Machine learning-enabled velocity model building with uncertainty quantification
• Bayesian Variational Time-lapse Full-waveform Inversion
• Elucidating the Design Space of Diffusion-Based Generative Models
• Conditional Image Generation with Score-Based Diffusion Models
• Score-Based Generative Modeling through Stochastic Differential Equations
• Denoising Diffusion Probabilistic Models
• A Simple Framework for Contrastive Learning of Visual Representations
• The frontier of simulation-based inference
• Contrastive Multiview Coding
• Contrastive Multiview Coding
• On Calibration of Modern Neural Networks

Cited by

• Dual-space posterior sampling for Bayesian inference in constrained inverse problems

Related

• Diffusion Model-Based Posterior Sampling in Full Waveform Inversion
• DiffusionInv: Prior-enhanced Bayesian Full Waveform Inversion using Diffusion models
• Machine learning-enabled velocity model building with uncertainty quantification
• Generative Geostatistical Modeling from Incomplete Well and Imaged Seismic Observations with Diffusion Models
• Stochastic full waveform inversion with deep generative prior for uncertainty quantification