1887

Abstract

Summary

In the past few years, the deep learning community has witnessed the explosive growth and rapid evolution of generative AI (GenAI) and large language models (LLMs). The scaling law stemming from the deepened network structures and accumulated data abundance has been consistently validated by different works in various industries. While works have been published discussing the potential of building large seismic FMs, the development of such a new paradigm for processing and interpreting seismic data is still in an explorative stage. In this paper, we aim at building a new paradigm of interpreting seismic images using a pretrained seismic foundation model. A reference-based geofeature extraction method is proposed to enable a more dynamic and heuristic seismic interpretation approach. With one-shot, or potentially few-shot labelling, a prompt-like interactive interpretation can be achieved. Unlike many of the current foundation model setups, the proposed geofeature extraction model is pretrained on a large amount of seismic data with both self-supervised and supervised approaches, and is used directly for inference without finetuning, ensuring efficiency and adaptiveness simultaneously.

© EAGE Publications BV
Loading

Article metrics loading...

/content/papers/10.3997/2214-4609.202510724
2025-06-02
2026-02-11

  • From This Site

    /content/papers/10.3997/2214-4609.202510724
    dcterms_title,dcterms_subject,pub_keyword
    -contentType:Journal -contentType:Contributor -contentType:Concept -contentType:Institution
    10
    5
Loading full text...

Full text loading...

References

  1. Chevitarese, D., Szwarcman, D., Silva, R.M.D. and Brazil, E.V. [2018]. Seismic facies segmentation using deep learning. AAPG Annual and Exhibition, 2018.
     [Google Scholar]
  2. Di, H., Wang, Z. and AlRegib, G. [2018]. Deep convolutional neural networks for seismic salt-body delineation. AAPG Annual Convention and Exhibition, 2018.
     [Google Scholar]
  3. Kirillov, A., Mintun, E., Ravi, N., Mao, H., Rolland, C., Gustafson, L., Xiao, T., Whitehead, S., Berg, A.C., Lo, W.Y. and Dollar, P. [2023]. Segment anything. Proceedings of the IEEE/CVF International Conference on Computer Vision, 4015–4026.
     [Google Scholar]
  4. KaiG and TingC. [2024], RGM: random geological model generation package.
     [Google Scholar]
  5. Ravi, N., Gabeur, V., Hu, Y.T., Hu, R., Ryali, C., Ma, T., Khedr, H., Radle, R., Rolland, C., Gustafson, L. and Mintun, E. [2024]. Sam 2: segment anything in images and videos. arXiv Preprint arXiv:2408.00714.
     [Google Scholar]
  6. Shi, Y., Wu, X. and Fomel, S. [2020]. Waveform embedding: Automatic horizon picking with unsupervised deep learning. Geophysics, 85(4), WA67–WA76.
     [Google Scholar]
  7. Sheng, H., Wu, X., Si, X., Li, J., Zhang, S. and Duan, X. [2023]. Seismic Foundation Model (SFM): a new generation deep learning model in geophysics. arXiv Preprint arXiv:2309.02791.
     [Google Scholar]
  8. Xiong, W., Ji, X., Ma, Y., Wang, Y., Al Bin Hassan, N.M., Ali, M.N. and Luo, Y. [2018]. Seismic fault detection with convolutional neural network. Geophysics, 83(5), 097–0103.
     [Google Scholar]
  9. Yuan, P., Wang, S., Hu, W., Wu, X., Chen, J. and Van Nguyen, H. [2020]. A robust first-arrival picking workflow using convolutional and recurrent neural networks. Geophysics, 85(5), U109–U119.
     [Google Scholar]
/content/papers/10.3997/2214-4609.202510724
Loading
Fast Reference-Based Geofeature Extraction by Seismic Foundation Model
Conference Proceedings 2025, 1 (2025); https://doi.org/10.3997/2214-4609.202510724
/content/papers/10.3997/2214-4609.202510724
/content/papers/10.3997/2214-4609.202510724
Loading

Data & Media loading...

Most Cited This Month Most Cited RSS feed

This is a required field
Please enter a valid email address
Approval was a Success
Invalid data
An Error Occurred
Approval was partially successful, following selected items could not be processed due to error