GeoLG-3DFaultNet

Labels fault surfaces voxel by voxel within a 3D seismic reflection volume.

Field
Geophysics
Method
Deep learning
Data it takes
volume
License
MIT

Fault interpretation in 3D seismic volumes is normally manual and slow. This classifies each voxel as fault or not, producing a fault probability volume that an interpreter can work from.

It fills a capability gap: nothing else in this catalog reads 3D seismic volumes.

Treat it as an evaluation candidate rather than an established tool. The method rests on a manuscript whose publication status we have not been able to confirm, so results should be checked against interpretation you trust before being relied on.

Upstream project →

Catalog entry last checked 2026-08-19. All models →

Run GeoLG-3DFaultNet on your data

The lab runs this on its own compute, in a container, with the inputs and parameters recorded alongside the result. Initial scoping conversations are free.

Contact the Lab