LLGF-Net: Learning Local and Global Feature Fusion for 3D Point Cloud Semantic Segmentation

Abstract

Three-dimensional (3D) point cloud semantic segmentation is fundamental in complex scene perception. Currently, although various efficient 3D semantic segmentation networks have been proposed, the overall effect has a certain gap to 2D image segmentation. Recently, some transformer-based methods have opened a new stage in computer vision, which also has accelerated the effective development of methods in 3D point cloud segmentation. In this paper, we propose a novel semantic segmentation network named LLGF-Net that can aggregate features from both local and global levels of point clouds, effectively improving the ability to extract feature information from point clouds. Specifically, we adopt the multi-head attention mechanism in the original Transformer model to obtain the local features of point clouds and then use the position-distance information of point clouds in 3D space to obtain the global features. Finally, the local features and global features are fused and embedded into the encoder–decoder network to generate our method. Our extensive experimental results on the 3D point cloud dataset demonstrate the effectiveness and superiority of our method.

Steel surface defect detection algorithm based on ESI-YOLOv8

“…To ensure a more objective comparison of the improved model, we analysed its performance alongside the current mainstream models, including YOLOX, Faster R-CNN, SSD,Resnet-50,YOLOv3, YOLOv4, YOLOv5, YOLOv7-tiny, and YOLOv8. Furthermore, we compared it with several other models, including the CBAM-MobilenetV2-YOLOv5 model (CM-YOLOv5) proposed by Yang [32] et al, the YOLO-ACG model by Wang [33] et al, the AGCN model by Zhang [34] et al, and the improved YOLOv8 model by Wei [35] et al, the multi-scale lightweight neural network model (MM) proposed by Shao [36] et al, and Zhang [37] et al proposed a model that combines CNN and Transformer. The experiments were conducted using identical hardware and software configurations, and the same dataset of steel surface defects was used.…”

Point Cloud Segmentation Network Based on Attention Mechanism and Dual Graph Convolution

“…At the same time, F in is fed into the convolution layer to generate a new feature vector F C ∈ R N×C , and then the similarity matrix S from Equation ( 7) is multiplied with F C , and the operation result is summed element-wise with F in to obtain the final output result F out . The calculation formula is shown in Equation (8).…”

“…In contrast, deep learning-based algorithms offer efficient computations and strong generalization capabilities that enable them to handle large-scale point clouds effectively. Consequently, they have gradually become dominant in the field of point cloud semantic segmentation [7,8]. In recent years, various deep learningbased segmentation networks have been proposed by researchers, which can be categorized into three types of methods: projection-based [9][10][11][12][13], voxelization-based [14][15][16][17][18], and pointbased [19][20][21][22][23].…”

A Discordance Analysis in Manual Labelling of Urban Mobile Laser Scanning Data Used for Deep Learning Based Semantic Segmentation

“…Although DL approaches do not yet achieve perfect segmentation results for complete scenarios, they have already achieved very high success rates on many tasks. For example, building floors were segmented with 98.5%, 98.2%, and 98.1% of IoU (Intersection over Union) using PointWeb (Zhao et al, 2019), LLGF-Net (Zhang et al, 2022), and HybridCR (M. Li et al, 2022) methods, respectively. Roads were segmented with 98.5% of IoU using CGA-Net (Lu et al, 2021), vegetation were segmented with 97.3% of IoU using MinkowskiNet (Choy et al, 2019), and buildings were segmented with 97.5% of F 1-Score using ResNet18 algorithms.…”