Denoising Diffusion Probabilistic Models for 2× Seismic Super-Resolution: Results from the Sleipner 4D CO₂ Monitoring Benchmark

Author(s)	Mr. Sunil Manikani, Dr. Murad Yaghi
Country	India
Abstract	We present a Denoising Diffusion Probabilistic Model (DDPM) approach to 2× seismic super-resolution (SR) applied to the Sleipner 4D CO2 monitoring benchmark (CO2DataShare/Equinor). A 2.73M-parameter UNet, conditioned on bicubic-upsampled low-resolution seismic inlines, learns to reverse a 100-step cosine-scheduled diffusion process to recover high-resolution inlines from Gaussian-blurred, 2× bicubic-downsampled observations. The model achieves a best validation PSNR of 34.56 dB and SSIM of 0.9216, compared to a bicubic baseline of 29.37 ± 0.87 dB and SSIM 0.7048 — a gain of +5.19 dB and +0.2168 SSIM. Visual inspection confirms substantially improved reflector continuity, wavelet fidelity, and amplitude contrast across geologically diverse inline sections of the Sleipner volume.
Keywords	Denoising, seismic, deep learning, diffusion, AI, ML, super resolution
Field	Computer > Artificial Intelligence / Simulation / Virtual Reality
Published In	Volume 7, Issue 5, September-October 2026
Published On	2026-09-29

About AIJFR Fees & Payment Current Issue Publication Archive	Submit Research Paper Track Submission Status Publication Guidelines Publication Ethics Peer Review & Plagiarism	Join as a Reviewer Editors & Reviewers Reviewer Referral Program Get Reviewer Membership Certi.	Website/Journal Policies Usage Policy Content Policies Privacy Policy

Contact Us		+91-9898-79-39-59	editor@aijfr.com

Advanced International Journal for Research