seismic-deeplearning and seismic-transfer-learning
About seismic-deeplearning
microsoft/seismic-deeplearning
Deep Learning for Seismic Imaging and Interpretation
Provides extensible ML pipelines with state-of-the-art segmentation models (UNet, SEResNET, HRNet) for facies classification and seismic facies segmentation on 2D/3D rectangular seismic volumes. Built on PyTorch/TensorFlow with modular, swappable model configurations and SEGY-to-numpy data conversion utilities. Integrates with Azure Machine Learning for distributed training and includes Docker containerization, Jupyter notebooks, and pip-installable `cv_lib` and `interpretation` utilities for end-to-end seismic workflows.
About seismic-transfer-learning
JesperDramsch/seismic-transfer-learning
Deep-learning seismic facies on state-of-the-art CNN architectures
This project helps geophysicists and exploration seismologists automatically identify different geological features, known as seismic facies, in 2D seismic survey images. By applying advanced image classification models pre-trained on vast photograph databases, it takes raw seismic sections and produces classified interpretations of subsurface structures. This allows specialists to quickly analyze and understand geological formations without extensive manual effort.
Related comparisons
Scores updated daily from GitHub, PyPI, and npm data. How scores work