An interactive framework integrating segment anything model and structure-from-motion for three-dimensional discontinuity identification in rock masses

• SAM achieves high accuracy in 2D discontinuity segmentation with 0.78 mean IoU and 0.86 average precision on a large dataset. • The proposed SAM-SfM framework extends identification to 3D, achieving 0.91 average precision on benchmark datasets. • The framework leverages the inherent pixel-point cloud relationship in SfM, simplifying and generalizing across photogrammetric devices. • The method provides a practical, efficient solution for rock mass discontinuity identification, valuable for geological investigations and data annotation.