发表机构
The University of Texas at Austin; Los Alamos National Laboratory; Uppsala University; The University of Texas at Permian Basin; King Fahd University of Petroleum and Minerals（德克萨斯大学奥斯汀分校; 洛斯阿拉莫斯国家实验室; 乌普萨拉大学; 德克萨斯大学佩米安盆地分校; 法赫德国王石油与矿产大学）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
GeoFWI3D是一个包含10,000个三维速度模型的开源基准数据集，旨在通过深度学习辅助地震成像和全波形反演，并提供多模态标签和基准任务以促进评估。
AI 中文摘要
我们推出了GeoFWI3D，这是一个大规模开源基准数据集，包含地质上合理的三维地下模型，旨在加速深度学习（DL）辅助的地震成像和全波形反演（FWI）。FWI是一种基于物理、以波动方程为基础的优化技术，用于地震成像中估计地下属性。然而，传统的FWI高度非线性、非唯一且不适定。当初始模型不够准确时，迭代过程往往收敛到局部最小值。最重要的是，三维FWI的计算成本极其高昂。随着深度学习的出现，可以通过在真实模型上训练网络来建立炮集与地下属性之间的直接映射。此类方法的主要瓶颈是缺乏大规模、真实的训练数据集。GeoFWI3D通过提供10,000个地质多样的速度模型（分辨率为96x96x96）弥补了这一空白，这些模型涵盖四个结构复杂度类别：纯地层、断层网络、盐底辟作用以及复杂的断层-盐耦合系统。每个模型都配有共同配准的多模态标签，包括纵波速度（Vp）、零偏移距地震反射率、相对地质时间（RGT）以及语义断层/盐体掩膜。为了促进系统性评估，我们引入了基准任务，涵盖断层检测、联合盐体分割与年代地层预测、FWI、使用神经算子的波场和走时代理模型，以及使用三维扩散模型的生成建模。我们为每项任务提供了基线结果，以建立供未来用户使用的参考性能指标。该数据集根据知识共享署名4.0国际许可公开提供。
英文摘要
We introduce GeoFWI3D, a large-scale open-source benchmark dataset of geologically plausible 3D subsurface models designed to accelerate deep learning (DL) assisted seismic imaging and full waveform inversion (FWI). FWI is a physics-driven, wave-equation-based optimization technique used in seismic imaging to estimate the subsurface properties. However, traditional FWI is highly non-linear, non-unique, and ill-posed. The iterative process often converges to local minima when the starting model is insufficiently accurate. Most importantly, 3D FWI is prohibitively expensive. With the advent of DL, a direct mapping between the shot gathers and subsurface properties can be established by training a network on realistic models. The primary bottleneck for such approaches is the lack of large-scale, realistic training datasets. GeoFWI3D addresses this gap with 10,000 geologically diverse velocity models at 96x96x96 resolution, spanning four structural complexity classes: pure stratigraphy, faulted networks, salt diapirism, and complex coupled fault-salt systems. Each model is accompanied by co-registered multi-modal labels including compressional velocity (Vp), zero-offset seismic reflectivity, relative geologic time (RGT), and semantic fault/salt masks. To facilitate systematic evaluation, we introduce benchmark tasks covering fault detection, joint salt body segmentation and chronostratigraphy prediction, FWI, wavefield and traveltime surrogates using neural operators, and generative modeling with a 3D diffusion model. We present baseline results for each task to establish reference performance metrics for future users. The dataset is publicly available under Creative Commons Attribution 4.0 International.
