Parallel Dual-Level Attention-Enhanced Semantic Segmentation of 3D Point Clouds in Tunnel Construction

Research on Point Cloud Segmentation Method for Ancient Building Interior Scenes Based on Geometric Feature Multi-Scale Network

Aiming at addressing the problems of complex geometric features, high inter-class similarity, and insufficient single-scale information in semantic segmentation of point clouds for ancient building interior components, this paper takes the Ba Wang Academy of Shenyang Jianzhu University as the research object and proposes a Geometric Feature Multi-scale Network (GFMN). First, a point cloud dataset containing five types of components—windows, beams, walls, roofs, and columns—was collected and constructed using a FARO Focus3D X330 terrestrial laser scanner (FARO Technologies, Lake Mary, Florida, USA). Second, 46-dimensional handcrafted geometric descriptors were extracted for each discrete point from four aspects: basic point attributes, local geometric features, density and scale features, and multi-scale fusion. On this basis, features were grouped according to semantics and fed into independent encoding branches, where a gated adaptive fusion mechanism was employed to dynamically adjust the contribution of each branch, and optimization was performed in combination with a prototype classification head and a joint loss function. Experimental results show that the proposed method achieved an overall accuracy of 93.17% on the test set, significantly outperforming state-of-the-art methods such as PointNet, PointNet++, Point Transformer, and Point Cloud Transformer. This study provides an effective solution for high-precision semantic segmentation of ancient building interior components.

Spmixnet: spatial-channel collaborative modeling for enhanced small-object segmentation in LiDAR point clouds

Unifying Street Scene Point Cloud Semantic Segmentation with Deformable Mesh-based Neural Representation

Abstract. Accurate semantic segmentation of urban point clouds is important for applications such as urban planning and autonomous driving. Recently, neural scene representations have been extended to merge semantic information across modalities and spatial dimensions. While 3D Gaussian Splatting (3DGS) enables efficient and high-quality reconstruction, its semantic understanding performance in street scenes is influenced by trajectory-constrained viewpoints, where Gaussian densification introduces occlusions and semantic ambiguity. This paper explores the use of NeRF-based neural representation for street scene point cloud semantic segmentation. Specifically, deformable neural mesh primitives (DNMPs) are used to compactly represent spatial geometry and simplify ray sampling. Then, neural fields including density, RGB, and semantics are constructed based on mesh vertex feature interpolation and MLPs. The sampled neural field values are accumulated via ray rendering and supervised using original images and corresponding semantic label maps generated by pre-trained models. Point cloud semantics are then predicted by interpolating neighboring samples within the learned field. The method is validated on the KITTI-360 and Waymo datasets. Results show that the proposed approach achieves improved semantic segmentation performance while maintaining competitive rendering quality, and supports both novel view synthesis and semantic rendering.

PDES-Net: LiDAR point cloud semantic segmentation network based on point-wise distance encoding and pointed-seg head

Light Detection and Ranging (LiDAR)-based semantic segmentation is significant in advanced autonomous driving systems. However, it is challenging to achieve accurate and efficient semantic segmentation because of the sparse and uneven distribution of LiDAR point cloud data. To address the above problem, we propose a LiDAR point cloud semantic segmentation network based on point-wise distance encoding and pointed-seg head, named as PDES-Net. The point-wise distance encoding mechanism introduces normalized continuous depth information as a supplementary feature into the network, enhances the perception of both near and distant points, and reduce geometric information loss. The pointed-seg head module adaptively integrates multilevel features through learnable weight coefficients, and enhances the expressive ability of point-wise prediction. The performance of the proposed PDES-Net is evaluated on the publicly available benchmarks, SemanticKITTI and nuScenes, achieving mIoU of 68.9% and 78.9%, respectively. The proposed PDES-Net enhances segmentation performance while maintaining advantages in model parameters and inference speed. Overall, the network achieves a well-balanced trade-off between accuracy and computational efficiency.

DCA-Net: Dilated Context Attention Network for MLS Point Cloud Semantic Segmentation

The proposed Dilated Context Attention Network (DCA-Net), which consists of a dilated local geometric encoding module, a channel attention pooling module, and a category-boundary sampling strategy, alleviates boundary confusion in point cloud segmentation with long-tail categories.

LERPNet: a lightweight and efficient range view point cloud semantic segmentation network

This paper replaces the Stem layer in FRNet with the proposed FD-Stem, which improves feature representation while reducing computational complexity, and introduces long-range modeling capability with limited additional parameters, enabling effective learning of both spatial and channel-wise representations.