cigFaciesNet is an open source code created by the Computational Interpretation Group (CIG) for "cigFacies: a massive-scale benchmark dataset of seismic facies and its application". Hui Gao, Xinming Wu, Xiaoming Sun and Mingcai Hou are the main contributors to the code.
This publicly available code contains three strategies of field seismic data curation, knowledge-guided synthesization and AI-based generation we proposed for building the benmark dataset, as well as training a network and then apply it for automatic siesmic facies classification on the validation datasets and two distinct 3-D field seismic data.
The benchmark dataset of seismic facies (cigFacies) constructed by these three strategies has been uploaded uploaded to Zenodo and are freely available at https://zenodo.org/records/10777460