Three-dimensional Unfolding and Unfaulting Dataset for Structural Interpretation Using Self-Supervised Learning

This dataset accompanies the study "Three-dimensional unfolding and unfaulting for structural interpretation using self-supervised learning." It provides synthetic data used to develop and validate a novel 3-D lightweight neural network framework for restoring deformed geological structures to their flattened state. The data include highly deformed examples with complex faulting and folding, designed to test the model's ability to compute precise shifts that realign stratigraphic layers.本数据集配套于题为《基于自监督学习的构造解释三维展开与去断层化（Three-dimensional unfolding and unfaulting for structural interpretation using self-supervised learning）》的研究，用于提供合成数据以开发并验证一款全新的三维轻量化神经网络框架，该框架可将变形的地质构造恢复至原始平整状态。数据集包含带有复杂断层与褶皱的高度变形样本，旨在测试模型计算精确位移以重新对齐地层的能力。