1887
Volume 74, Issue 2
  • E-ISSN: 1365-2478

Abstract

ABSTRACT

Full‐waveform inversion (FWI) is a standard velocity model‐building (VMB) technique in seismic exploration, yet conventional implementations struggle to deliver high‐resolution three‐dimensional (3D) velocity models due to strong dependencies on accurate initial models – frequently unavailable in field applications. While data‐driven deep learning solutions show potential, they face significant challenges including prohibitive training data requirements, computational burdens and limited generalization. To overcome these limitations, we propose a low‐rank adaptive U‐Net with attention gates (LAUNet): a novel foundation‐model‐inspired two‐dimensional (2D) U‐Net framework integrating enhanced low‐rank adaptation (LoRA) and attention gates, enabling parameter‐efficient fine‐tuning for 3D seismic VMB. The method processes 2D initial velocity slices and the corresponding reverse time migration slices extracted from 3D volumes, leveraging self‐attention to fuse multimodal features for enhanced VMB precision. This approach resolves dimensionality inconsistencies inherent to acquisition geometries while reducing computational costs compared to 3D FWI. Additionally, we introduce an adaptive LoRA fine‐tuning strategy that incorporates well‐log data. By leveraging pre‐trained representations and multimodal inputs, LAUNet achieves high fidelity in velocity prediction with minimal labelled data, demonstrating strong generalization across complex geological scenarios. Experimental validation on 3D Overthrust and SEAM models demonstrates LAUNet's superiority over traditional FWI. Numerical results confirm enhanced inversion accuracy in deep, complex geological structures alongside improved computational efficiency and generalization capability.

© 2026 European Association of Geoscientists & Engineers. © 2026 European Association of Geoscientists & Engineers.
Loading

Article metrics loading...

/content/journals/10.1111/1365-2478.70134
2026-02-15
2026-03-13

  • From This Site

    /content/journals/10.1111/1365-2478.70134
    dcterms_title,dcterms_subject,pub_keyword
    -contentType:Journal -contentType:Contributor -contentType:Concept -contentType:Institution
    10
    5
Loading full text...

Full text loading...

References

  1. Al‐Yahya, K1989. “Velocity Analysis By Iterative Profile Migration.” Geophysics54, no. 6: 718–729.
     [Google Scholar]
  2. Baysal, E., D. D.Kosloff, and J. W.Sherwood. 1983. “Reverse Time Migration.” Geophysics48, no. 11: 1514–1524.
     [Google Scholar]
  3. Bunks, C., F. M.Saleck, S.Zaleski, and G.Chavent. 1995. “Multiscale Seismic Waveform Inversion.” Geophysics60, no. 5: 1457–1473.
     [Google Scholar]
  4. Chattopadhyay, S., and G. A.McMechan. 2008. “Imaging Conditions for Prestack Reverse‐Time Migration.” Geophysics73, no. 3: S81–S89.
     [Google Scholar]
  5. Chiao, L., and B.Kuo. 2001. “Multiscale Seismic Tomography.” Geophysical Journal International145, no. 2: 517–527.
     [Google Scholar]
  6. Engquist, B., B. D.Froese, and Y.Yang. 2016. “Optimal Transport for Seismic Full Waveform Inversion.” Communications in Mathematical Sciences14, no. 8: 2309–2330.
     [Google Scholar]
  7. Gao, H., X.Wu, and L.Liang, et al. 2025. “A Foundation Model Empowered by a Multi‐Modal Prompt Engine for Universal Seismic Geobody Interpretation Across Surveys.” Information Fusion125: 103437.
     [Google Scholar]
  8. Gauthier, O., J.Virieux, and A.Tarantola. 1986. “Two‐Dimensional Nonlinear Inversion of Seismic Waveforms: Numerical Results.” Geophysics51, no. 7: 1387–1403.
     [Google Scholar]
  9. Gelboim, M., A.Adler, Y.Sun, and M.Araya‐Polo. 2022. “Encoder–Decoder Architecture for 3D Seismic Inversion.” Sensors23, no. 1: 61.
     [Google Scholar]
  10. Gelboim, M., A.Adler, Y.Sun, and M.Araya‐Polo. 2023. “Deep Compressed Learning for 3D Seismic Inversion.” Paper presented at the SEG International Exposition and Annual Meeting, SEG–2023.
  11. Geng, Z., Z.Zhao, Y.Shi, X.Wu, S.Fomel, and M.Sen. 2022. “Deep Learning for Velocity Model Building With Common‐Image Gather Volumes.” Geophysical Journal International228, no. 2: 1054–1070.
     [Google Scholar]
  12. Guo, Z., X.Wu, L.Liang, H.Sheng, N.Chen, and Z.Bi. 2025. “Cross‐Domain Foundation Model Adaptation: Pioneering Computer Vision Models for Geophysical Data Analysis.” Journal of Geophysical Research: Machine Learning and Computation2, no. 1: e2025JH000601.
     [Google Scholar]
  13. Han, K., Y.Wang, and H.Chen, et al. 2022. “A Survey on Vision Transformer.” IEEE Transactions on Pattern Analysis and Machine Intelligence45, no. 1: 87–110.
     [Google Scholar]
  14. Han, L., X.Wu, Z.Hu, J.Li, and H.Fang. 2024. “MAMCL: Multi‐Attributes Masking Contrastive Learning for Explainable Seismic Facies Analysis.” Computers & Geosciences193: 105731.
     [Google Scholar]
  15. Harsuko, R., S.Cheng, and T.Alkhalifah. 2025. “Propagating the Prior From Shallow to Deep With a Pre‐Trained Velocity‐Model Generative Transformer Network.” Journal of Geophysical Research: Machine Learning and Computation2, no. 1: e2024JH000408.
     [Google Scholar]
  16. He, Q., and Y.Wang. 2021. “Reparameterized Full‐Waveform Inversion Using Deep Neural Networks.” Geophysics86, no. 1: V1–V13.
     [Google Scholar]
  17. Helbig, K. 1994. Foundations of Anisotropy for Exploration Seismics. Pergamon Press.
     [Google Scholar]
  18. Houlsby, N., A.Giurgiu, and S.Jastrzebski, et al. 2019. “Parameter‐Efficient Transfer Learning for NLP.” In International Conference on Machine Learning, 2790–2799. PMLR,Cambridge, MA.
     [Google Scholar]
  19. Hu, E. J., Y.Shen, and P.Wallis, et al. 2022. “LoRA: Low‐Rank Adaptation of Large Language Models.” ICLR1, no. 2: 3.
     [Google Scholar]
  20. Huang, X., and S.Belongie. 2017. “Arbitrary Style Transfer in Real‐Time With Adaptive Instance Normalization.” In Proceedings of the IEEE International Conference on Computer Vision1501–1510. IEEE.
     [Google Scholar]
  21. Jia, M., L.Tang, and B.‐C.Chen, et al. 2022. “Visual Prompt Tuning.” In European Conference on Computer Vision, 709–727. Springer.
     [Google Scholar]
  22. Jones, I. F.2010. An Introduction to: Velocity Model Building. EAGE.
     [Google Scholar]
  23. Khosro Anjom, F., F.Vaccarino, and L. V.Socco. 2023. “Machine Learning for Seismic Exploration: Where Are We and How Far Are We from the Holy Grail?.” Geophysics88, no. 6: 1111–1403.
     [Google Scholar]
  24. Li, S., B.Liu, and Y.Ren, et al. 2020. “Deep‐Learning Inversion of Seismic Data.” IEEE Transactions on Geoscience and Remote Sensing58, no. 3: 2135–2149.
     [Google Scholar]
  25. Li, Y., T.Alkhalifah, J.Huang, and Z.Li. 2023. “Self‐Supervised Pretraining Vision Transformer With Masked Autoencoders for Building Subsurface Model.” IEEE Transactions on Geoscience and Remote Sensing61: 1–10.
     [Google Scholar]
  26. Li, Y., T.Alkhalifah, and Z.Zhang. 2021. “Deep‐Learning Assisted Regularized Elastic Full Waveform Inversion Using the Velocity Distribution Information From Wells.” Geophysical Journal International226, no. 2: 1322–1335.
     [Google Scholar]
  27. Li, Y. E., and L.Demanet. 2016. “Full‐Waveform Inversion With Extrapolated Low‐Frequency Data.” Geophysics81, no. 6: R339–R348.
     [Google Scholar]
  28. Lin, Y., J.Theiler, and B.Wohlberg. 2023. “Physics‐Guided Data‐Driven Seismic Inversion: Recent Progress and Future Opportunities in Full‐Waveform Inversion.” IEEE Signal Processing Magazine40, no. 1: 115–133.
     [Google Scholar]
  29. Liu, Q., Z.Chen, and J.Ma. 2025. “Foundation Models for Exploration Geophysics.” Geophysics90, no. 6: H29–H52.
     [Google Scholar]
  30. Louboutin, M., M.Lange, and F.Luporini, et al. 2019. “Devito (v3. 1.0). An Embedded Domain‐Specific Language for Finite Differences and Geophysical Exploration.” Geoscientific Model Development12, no. 3: 1165–1187.
     [Google Scholar]
  31. Moor, M., O.Banerjee, and Z. S. H.Abad, et al. 2023. “Foundation Models for Generalist Medical Artificial Intelligence.” Nature616, no. 7956: 259–265.
     [Google Scholar]
  32. Oktay, O., J.Schlemper, and L. L.Folgoc, et al. 2018. “Attention U‐Net: Learning Where to Look for the Pancreas.” Preprint arXiv, April 18. https://doi.org/10.48550/arXiv.1804.03999.
  33. Ordonez, A., D.Wade, C.Ravaut, and A.Waldeland. 2024. “Towards a Foundation Model for Seismic Interpretation.” In 85th EAGE Annual Conference & Exhibition, vol. 2024, 1–5. European Association of Geoscientists & Engineers.
     [Google Scholar]
  34. Pasasa, L., F.Wenzel, and P.Zhao. 1998. “Prestack Kirchhoff Depth Migration of Shallow Seismic Data.” Geophysics63, no. 4: 1241–1247.
     [Google Scholar]
  35. Paszke, A., S.Gross, and F.Massa, et al. 2019. “Pytorch: An Imperative Style, High‐Performance Deep Learning Library.” Advances in Neural Information Processing Systems32: 8026–8037.
     [Google Scholar]
  36. Plessix, R.‐E2006. “A Review of the Adjoint‐State Method for Computing the Gradient of a Functional with Geophysical Applications.” Geophysical Journal International167, no. 2: 495–503.
     [Google Scholar]
  37. Ren, Y., L.Nie, S.Yang, P.Jiang, and Y.Chen. 2021. “Building Complex Seismic Velocity Models for Deep Learning Inversion.” IEEE Access9: 63767–63778.
     [Google Scholar]
  38. Ronneberger, O., P.Fischer, and T.Brox. 2015. “U‐Net: Convolutional Networks for Biomedical Image Segmentation.” In International Conference on Medical Image Computing and Computer‐Assisted Intervention, 234–241. Springer.
     [Google Scholar]
  39. Sheng, H., X.Wu, X.Si, J.Li, S.Zhang, and X.Duan. 2025. “Seismic Foundation Model: A Next Generation Deep‐Learning Model in Geophysics.” Geophysics90, no. 2: IM59–IM79.
     [Google Scholar]
  40. Sun, B., and T.Alkhalifah. 2020. “ML‐Descent: An Optimization Algorithm for Full‐Waveform Inversion Using Machine Learning.” Geophysics85, no. 6: R477–R492.
     [Google Scholar]
  41. Sun, J., K.Innanen, T.Zhang, and D.Trad. 2023. “Implicit Seismic Full Waveform Inversion With Deep Neural Representation.” Journal of Geophysical Research: Solid Earth128, no. 3: e2022JB025964.
     [Google Scholar]
  42. W Symes, W.2008. “Migration Velocity Analysis and Waveform Inversion.” Geophysical Prospecting56, no. 6: 765–790.
     [Google Scholar]
  43. Virieux, J., and S.Operto. 2009. “An Overview of Full‐Waveform Inversion in Exploration Geophysics.” Geophysics74, no. 6: WCC1–WCC26.
     [Google Scholar]
  44. Wang, F., T.Alkhalifah, and X.Huang. 2024. “A Prior Regularized 3D Full‐Waveform Inversion Using 2D Generative Diffusion Models.” In Fourth International Meeting for Applied Geoscience & Energy, 988–992. Society of Exploration Geophysicists and American Association of Petroleum Geologists.
  45. Wang, H., J.Lin, X.Dong, S.Lu, Y.Li, and B.Yang. 2023. “Seismic Velocity Inversion Transformer.” Geophysics88, no. 4: R513–R533.
     [Google Scholar]
  46. Wang, W., G. A.McMechan, and J.Ma. 2021. “Elastic Isotropic and Anisotropic Full‐Waveform Inversions Using Automatic Differentiation for Gradient Calculations in a Framework of Recurrent Neural Networks.” Geophysics86, no. 6: R795–R810.
     [Google Scholar]
  47. Wu, Y., and Y.Lin. 2019. “InversionNet: An Efficient and Accurate Data‐Driven Full Waveform Inversion.” IEEE Transactions on Computational Imaging6: 419–433.
     [Google Scholar]
  48. Wu, Y., and G. A.McMechan. 2019. “Parametric Convolutional Neural Network‐Domain Full‐Waveform Inversion.” Geophysics84, no. 6: R881–R896.
     [Google Scholar]
  49. Yang, F., and J.Ma. 2019. “Deep‐Learning Inversion: A Next‐Generation Seismic Velocity Model Building Method.” Geophysics84, no. 4: R583–R599.
     [Google Scholar]
  50. Yang, F., and J.Ma. 2023a. “FWIGAN: Full‐Waveform Inversion Via a Physics‐Informed Generative Adversarial Network.” Journal of Geophysical Research: Solid Earth128, no. 4: e2022JB025493.
     [Google Scholar]
  51. Yang, F., and J.Ma. 2023b. “Wasserstein Distance‐Based Full‐Waveform Inversion With a Regularizer Powered by Learned Gradient.” IEEE Transactions on Geoscience and Remote Sensing61: 1–13.
     [Google Scholar]
  52. Yang, F., and J.Ma. 2025. “Gabor Wavelet‐Activation Implicit Neural Learning for Full‐Waveform Inversion.” Geophysics90, no. 3: R71–R87.
     [Google Scholar]
  53. Yang, L., S.Fomel, and S.Wang, et al. 2025. “HCTNet: Robust Prestack Seismic Inversion Using a Hybrid Convolutional Transformer.” Geophysics90, no. 4: N17–N32.
     [Google Scholar]
  54. Yang, S., T.Alkhalifah, Y.Ren, B.Liu, Y.Li, and P.Jiang. 2023. “Well‐Log Information‐Assisted High‐Resolution Waveform Inversion Based on Deep Learning.” IEEE Geoscience and Remote Sensing Letters20: 1–5.
     [Google Scholar]
  55. Yenduri, G., M.Ramalingam, and G. C.Selvi, et al. 2024. “GPT (Generative Pre‐Trained Transformer) – A Comprehensive Review on Enabling Technologies, Potential Applications, Emerging Challenges, and Future Directions.” IEEE Access12: 54608–54649.
     [Google Scholar]
  56. Yu, S., and J.Ma. 2021. “Deep Learning for Geophysics: Current and Future Trends.” Reviews of Geophysics59, no. 3: e2021RG000742.
     [Google Scholar]
  57. Zeng, Q., S.Feng, B.Wohlberg, and Y.Lin. 2021. “InversionNet3D: Efficient and Scalable Learning for 3‐D Full‐Waveform Inversion.” IEEE Transactions on Geoscience and Remote Sensing60: 1–16.
     [Google Scholar]
  58. Zhang, J., and M. N.Toksöz. 1998. “Nonlinear Refraction Traveltime Tomography.” Geophysics63, no. 5: 1726–1737.
     [Google Scholar]
  59. Zheng, H., L.Shen, and A.Tang, et al. 2025. “Learning From Models Beyond Fine‐Tuning.” Nature Machine Intelligence7, no. 1: 6–17.
     [Google Scholar]
  60. Zhou, Y., M. A.Chia, and S. K.Wagner, et al. 2023. “A Foundation Model for Generalizable Disease Detection From Retinal Images.” Nature622, no. 7981: 156–163.
     [Google Scholar]
  61. Zhu, W., K.Xu, E.Darve, B.Biondi, and G. C.Beroza. 2022. “Integrating Deep Neural Networks With Full‐Waveform Inversion: Reparameterization, Regularization, and Uncertainty Quantification.” Geophysics87, no. 1: R93–R109.
     [Google Scholar]
/content/journals/10.1111/1365-2478.70134
Loading
Towards Foundation Model‐Driven Geophysical Inversion by Low‐Rank Adaptation and Attention Gates
Geophysical Prospecting 74, e70134 (2026); https://doi.org/10.1111/1365-2478.70134
/content/journals/10.1111/1365-2478.70134
/content/journals/10.1111/1365-2478.70134
Loading

Data & Media loading...

  • Article Type: Research Article
Keyword(s): attention gates; deep learning; foundation model; low‐rank adaptation; parameter‐efficient fine‐tuning; three‐dimensional velocity model building

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