Summary of Seisfusion: Constrained Diffusion Model with Input Guidance For 3d Seismic Data Interpolation and Reconstruction, by Shuang Wang et al.
SeisFusion: Constrained Diffusion Model with Input Guidance for 3D Seismic Data Interpolation and Reconstruction
by Shuang Wang, Fei Deng, Peifan Jiang, Zishan Gong, Xiaolin Wei, Yuqing Wang
First submitted to arxiv on: 18 Mar 2024
Categories
- Main: Machine Learning (cs.LG)
- Secondary: Geophysics (physics.geo-ph)
GrooveSquid.com Paper Summaries
GrooveSquid.com’s goal is to make artificial intelligence research accessible by summarizing AI papers in simpler terms. Each summary below covers the same AI paper, written at different levels of difficulty. The medium difficulty and low difficulty versions are original summaries written by GrooveSquid.com, while the high difficulty version is the paper’s original abstract. Feel free to learn from the version that suits you best!
| Summary difficulty | Written by | Summary |
|---|---|---|
| High | Paper authors | High Difficulty Summary Read the original abstract here |
| Medium | GrooveSquid.com (original content) | Medium Difficulty Summary This paper proposes a novel diffusion model reconstruction framework for 3D seismic data, addressing the challenge of complex missing patterns in traditional methods. The framework introduces conditional supervision constraints and a 3D neural network architecture to generate reconstructions with higher consistency. Ablation studies determine optimal parameter values, demonstrating superior reconstruction accuracy on both field datasets and synthetic datasets. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary Seismic data is crucial for understanding earthquakes, but often has missing parts. This makes it hard to recreate the full data. Deep learning models can help, but they need special training for seismic data. The authors propose a new way to use diffusion models for 3D seismic data reconstruction. They add extra constraints to make sure the generated data is correct and incorporate missing data into the process. Their method works well on real-world and fake datasets with different types of missing patterns. |
Keywords
* Artificial intelligence * Deep learning * Diffusion model * Neural network
