2020 Sr Seg
Published in , 1900
Deep learning for simultaneous seismic super-resolution and denoising
Jintao Li, Xinming Wu and Zhanxuan Hu
SEG International Exposition and Annual Meeting 2020 (EI)
Paper ｜ Code 
Posts by Collection
publications
2024 01 Cigvis Eage
Published in , 1900
CIGVis: an open-source Python tool for real-time interactive visualization of multidimensional geophysical data
Jintao Li, Yunzhi Shi and Xinming Wu
85th EAGE Annual Conference & Exhibition, Jun 2024 (EI)
Paper | Code 
2024 02 Sfm Eage
Published in , 1900
Seismic Foundation Model (SFM): All-Purpose Feature Extraction from Seismic Data for Diverse Geophysical Applications
Hanlin Sheng, Xinming Wu, Xu Si, Jintao Li, Sibo Zhang, Xudong Duan
85th EAGE Annual Conference & Exhibition, Jun 2024 (EI)
Paper | Code 
2024 03 Si Agu
Published in , 1900
Tuning into Earth’s Voice: Leveraging Large Language Models for Earthquake Detection
Xu Si, Xinming Wu, Jintao Li, Xin Cui, Hanlin Sheng, Hang Gao, Zefeng Li, Zhigang Peng
AGU Fall Meeting Abstracts 2024
Paper
2025 03 Fm Review
Published in , 1900
On the workflow, opportunities and challenges of developing foundation model in geophysics
Hanlin Sheng, Xinming Wu, Hang Gao, Haibin Di, Sergey Fomel, Jintao Li, Xu Si
Review Paper | arXiv
2025 04 Dou
Published in , 1900
Geological Everything Model 3D: A Promptable Foundation Model for Unified and Zero-Shot Subsurface Understanding
Yimin Dou, Xinming Wu, Nathan L Bangs, Harpreet Singh Sethi, Jintao Li, Hang Gao, Zhixiang Guo
Under Revision | arXiv | project | code
2026 03 Reflectivity Geophysics
Published in , 1900
Deep Learning-based Seismic Reflectivity Estimation using Labeled Synthetic Data Pre-training and Physics-guided Fine-tuning in Field Data
Yuting Wang, Jintao Li†, Xiaoming Sun and Xinming Wu
Geophysics, Moderate revision
2026 04 Han
Published in , 1900
A Universal Diffusion Regularizer for Geophysical Inversion
Long Han, Xinming Wu†, Xiaoming Sun, Jintao Li, Yimin Dou
Submitted
2022 Sr Ieee Tgrs
Published in , 1900
Deep learning for simultaneous seismic super-resolution and denoising
Jintao Li, Xinming Wu and Zhanxuan Hu [ESI Highly Cited Paper!]
IEEE Transactions on Geoscience & Remote Sensing (IEEE TGRS), Vol. 60, pp. 1-11. 2022
Paper ｜ Code
2023 Facies Geophysics
Published in , 1900
Unsupervised contrastive learning for seismic facies characterization
Jintao Li, Xinming Wu, Yueming Ye, Cun Yang, Zhanxuan Hu, Xiaoming Sun and Tao Zhao
Geophysics, Vol. 88(1), WA81–WA89. 2023
Paper ｜ PDF
2024 Mamcl Cg
Published in , 1900
MAMCL: Multi-attributes Masking Contrastive Learning for explainable seismic facies analysis
Long Han, Xinming Wu, Zhanxuan Hu, Jintao Li, Huijing Fang
Computers & Geosciences, 193, p.105731. 2024
Paper
2025 01 Cigvis Geophysics
Published in , 1900
CIGVis: an open-source Python tool for real-time interactive visualization of multidimensional geophysical data
Jintao Li, Yunzhi Shi and Xinming Wu
Geophysics, 90(1), pp.1‑37. 2025
Paper | Code
2025 02 Sfm Geophysics
Published in , 1900
Seismic foundation model: A next generation deep-learning model in geophysics
Hanlin Sheng, Xinming Wu, Xu Si, Jintao Li, Sibo Zhang, Xudong Duan
Geophysics, 90(2), pp.IM59-IM79. 2025
Paper | Code
2026 01 Sr Tip
Published in , 1900
High‑Fidelity Seismic Super‑Resolution Using Prior‑Informed Deep Learning with 3D Awareness
Jintao Li, Xinming Wu, Xianwen Zhang, Xin Du, Xiaoming Sun, Bao Deng and Guangyu Wang
IEEE Transactions on Image Processing (IEEE TIP) 2026 [CCF-A, IF 13.7!]
Paper | [Code] | PDF
2026 02 Infer Ce
Published in , 1900
Memory-Efficient Full-Volume Inference for Large-Scale 3D Dense Prediction without Performance Degradation
Jintao Li and Xinming Wu
Communications Engineering 2026 [Nature Portfolio]
This paper was unanimously recognized by all three named reviewers
Paper | Code 
research
CIGVis: an open-source Python tool for real-time interactive visualization of multidimensional geophysical data Permalink
As Python’s role in processing and interpreting geophysical data expands, the need for a Python-based tool tailored for visualizing geophysical data has become increasingly critical. In response, CIGVis was developed - a fully open-source Python tool optimized for researchers and licensed under the MIT. It specializes in real-time, interactive visualiza- tion of multidimensional geophysical data, including 3D seismic, faults, horizons, geological bodies, and well logs. CIGVis enables users to interact with data through operations such as rotation, movement, zooming, and dragging slices. It also supports a multi-canvas functionality, allowing simultaneous visualization across multiple sub-canvases with a unified camera perspective. Its ease of use allows for effective visualization with just a few lines of code across all major operating systems, and extends to both desktop and Jupyter environments, facilitating code execution in various settings. The core functionalities of CIGVis are exemplified using straightforward datasets, such as the F3 dataset.
Memory-Efficient Full-Volume Inference for Large-Scale 3D Dense Prediction without Performance Degradation Permalink
Large-volume 3D dense prediction is essential in industrial applications such as energy exploration and medical image segmentation. However, existing deep learning models struggle to process full-size volumetric inputs at inference due to memory constraints and inefficient operator execution. Conventional solutions—such as tiling or compression—often introduce artifacts, compromise spatial consistency, or require retraining. We present a retraining-free inference optimization framework that enables accurate, efficient, whole-volume prediction without any performance degradation. Our approach integrates operator spatial tiling, operator fusion, normalization statistic aggregation, and on-demand feature recomputation to reduce memory usage and accelerate runtime. Validated across multiple seismic exploration models, our framework supports full size inference on volumes exceeding \(1024^3\) voxels. On FaultSeg3D, for instance, it completes inference on a \(1024^3\) volume in 7.5 seconds using just 27.6 GB of memory—compared to conventional inference, which can handle only \(448^3\) inputs under the same budget, marking a \(13\times\) increase in volume size without loss in performance. Unlike traditional patch-wise inference, our method preserves global structural coherence, making it particularly suited for tasks inherently incompatible with chunked processing, such as implicit geological structure estimation. This work offers a generalizable, engineering-friendly solution for deploying 3D models at scale across industrial domains.
High‑Fidelity Seismic Super‑Resolution Using Prior‑Informed Deep Learning with 3D Awareness Permalink
The limitations of seismic vertical resolution pose significant challenges for the identification of thin beds. Improving the vertical resolution of seismic data using deep learning methods often encounters challenges related to unrealistic outputs and limited generalization. To address these challenges, we propose a novel framework that improves the fidelity and generalization of seismic super-resolution. Our approach begins with the generation of realistic synthetic training data that aligns with the structural and amplitude characteristics of field surveys. We then introduce an enhanced 2D network with 3D awareness, which builds on the 2D Swin-Transformer and 3D convolution blocks to effectively capture 3D spatial features while maintaining computational efficiency. This network addresses the limitations of traditional 2D approaches by reducing stitching artifacts and improving spatial consistency. Finally, we develop a prior-informed fine-tuning strategy using field data without the need for labels, which incorporates a self-supervised data consistency loss and a spectral matching loss based on prior knowledge. This strategy ensures that the super-resolution results preserve the original low frequency information while yielding a spectral distribution as expected. Experiments on multiple field datasets demonstrate the robustness and generalization capability of our method, making it a practical solution for seismic resolution enhancement in diverse field datasets.