Large-Scale ALS Point Cloud Segmentation via Projection-Based Context Embedding.

Semantic segmentation of airborne laser scanning (ALS) point clouds is a valuable yet challenging task in remote sensing. When processing large-scale ALS scenes, it is necessary to partition them into smaller blocks for ease of handling. However, this partitioning introduces a challenge in capturing the ample spatial context within each block to adequately recognize the objects with a significant spatial span. This limitation becomes particularly pronounced when relying solely on the 3D representations as the input of nerual networks. To incorporate sufficient contextual information in ALS data semantic segmentation, we propose a multi-modal-based segmentation framework called projection-based context embedding (PCE) in this study. PCE effectively combines the advantages of 2D image and 3D point-voxel representations, which are the computational efficiency and the representation capability for fine-grained 3D geometries. The 2D projection is used to encode a large-scale semantic context, which is computationally expensive to be obtained using only pure 3D representation. Simultaneously, the sparse-point-voxel convolution (SPVConv) is employed to focus on learning 3D features from a small block of points centered on the large-scale context. Finally, to fully exploit the power of each modality, the embedding disentangling (ED) strategy is proposed additionally to combine the context embedding from the 2D image with 3D features for the final prediction. We demonstrate the state-of-the-art performance of PCE through extensive experiments on public large-scale ALS point cloud datasets.