Fig. 7From: Deep learning based approaches from semantic point clouds to semantic BIM models for heritage digital twinIllustration of our network architectures for segmentation of 3D point clouds. SFE, GFA represent our spatial feature enrichment module and global feature aggregation layer, respectively. During the forward pass, the input features are processed through a series of operations (each represented by a different colored box), while the points are passed to each layer as a structural guide for the operationsBack to article page