A Dual Attention KPConv Network Combined With Attention Gates for Semantic Segmentation of ALS Point Clouds

Kernel point convolution (KPConv) defines convolutional weights based on Euclidean distances between kernel points and input points and has shown good segmentation results on several datasets. However, it does not consider the intrinsic connection between input points and features, which is crucial for the semantic segmentation of airborne laser scanning (ALS) point clouds with sparse density and complex backgrounds. To address this problem, we design a dual attention KPConv network (DAKAG-Net) combined with attention gates for semantic segmentation of ALS point clouds. Specifically, we design the channel and spatial attention KPConv (CSAKPConv) block in the encoding process, which first performs adaptive feature refinement of the input mapping along two separate dimensions, channel and spatial, and then performs kernel point convolution. In addition, to enhance the use of high-level semantic information and detect objects of varying sizes, DAKAG-Net incorporates multiple attention gates (MAGs) that merge the lowest-level features, skip-connected features, and corresponding upsampled features during the decoding process. The decoded features are ultimately convolved with convolution kernels of various sizes and then merged to acquire multiscale perceptual field features. The proposed DAKAG-Net improves the OA, mF1, and mIoU by 3.5%, 3.1%, and 3.5%, respectively, compared with the baseline results on the ISPRS 3-D dataset, and yields the segmentation accuracy rates of 85.2% (OA), 73.7% (mF1), and 61.2% (mIoU). Moreover, the DAKAG-Net also obtains new state-of-the-art segmentation results on the DFC2019 dataset and the LASDU dataset.

• Publication year

  2024

• Venue

  IEEE Transactions on Geoscience and Remote Sensing

• Publication date

  Unknown publication date

• Fields of study

  Computer Science, Engineering, Environmental Science

• Identifiers
  DOI 10.1109/TGRS.2024.3422829
• External record

  Open on Semantic Scholar

• Source metadata

  Semantic Scholar

• No claims are published for this paper.

• No concepts are published for this paper.

• Extracting 3-D Structural Lines of Building From ALS Point Clouds Using Graph Neural Network Embedded With Corner Information

  2023cited by this paper
• Developing a new red band–SEVI–blue band (RSB) enhancement method for recognition the extra-high-voltage transmission line corridor in green mountains

  2023cited by this paper
• A tree detection method based on trunk point cloud section in dense plantation forest using drone LiDAR data

  2023cited by this paper
• MV-DeepSDF: Implicit Modeling with Multi-Sweep Point Clouds for 3D Vehicle Reconstruction in Autonomous Driving

  2023cited by this paper
• Prototype-Guided Multitask Adversarial Network for Cross-Domain LiDAR Point Clouds Semantic Segmentation

  2023cited by this paper
• OFFS-Net: Optimal Feature Fusion-Based Spectral Information Network for Airborne Point Cloud Classification

  2023cited by this paper
• Multilevel Context Feature Fusion for Semantic Segmentation of ALS Point Cloud

  2023cited by this paper
• Recurrent Residual Dual Attention Network for Airborne Laser Scanning Point Cloud Semantic Segmentation

  2023influential reference
• Power Line-Guided Automatic Electric Transmission Line Inspection System

  2022cited by this paper
• Beyond single receptive field: A receptive field fusion-and-stratification network for airborne laser scanning point cloud classification

  2022influential reference
• Local and Global Structure for Urban ALS Point Cloud Semantic Segmentation with Ground-Aware Attention

  2022cited by this paper
• High-quality indoor scene 3D reconstruction with RGB-D cameras: A brief review

  2022cited by this paper
• PointFormer: A Dual Perception Attention-Based Network for Point Cloud Classification

  2022cited by this paper
• VD-LAB: A view-decoupled network with local-global aggregation bridge for airborne laser scanning point cloud classification

  2022cited by this paper
• DenseKPNET: Dense Kernel Point Convolutional Neural Networks for Point Cloud Semantic Segmentation

  2022cited by this paper
• CSF-Net: Color Spectrum Fusion Network for Semantic Labeling of Airborne Laser Scanning Point Cloud

  2022cited by this paper
• A Dual Attention Neural Network for Airborne LiDAR Point Cloud Semantic Segmentation

  2022cited by this paper
• Airborne LiDAR point cloud classification with global-local graph attention convolution neural network

  2021cited by this paper
• A method for organs classification and fruit counting on pomegranate trees based on multi-features fusion and support vector machine by 3D point cloud

  2020cited by this paper
• LASDU: A Large-Scale Aerial LiDAR Dataset for Semantic Labeling in Dense Urban Areas

  2020influential reference
• Deep point embedding for urban classification using ALS point clouds: A new perspective from local to global

  2020cited by this paper
• A Closer Look at Local Aggregation Operators in Point Cloud Analysis

  2020cited by this paper
• Relation-Shape Convolutional Neural Network for Point Cloud Analysis

  2019cited by this paper
• RandLA-Net: Efficient Semantic Segmentation of Large-Scale Point Clouds

  2019cited by this paper
• Directionally Constrained Fully Convolutional Neural Network For Airborne Lidar Point Cloud Classification

  2019cited by this paper
• KPConv: Flexible and Deformable Convolution for Point Clouds

  2019influential reference
• Semantic Stereo for Incidental Satellite Images

  2018influential reference
• Classifying airborne LiDAR point clouds via deep features learned by a multi-scale convolutional neural network

  2018influential reference
• Deep Parametric Continuous Convolutional Neural Networks

  2018cited by this paper
• CBAM: Convolutional Block Attention Module

  2018cited by this paper
• VV-Net: Voxel VAE Net With Group Convolutions for Point Cloud Segmentation

  2018cited by this paper
• Dynamic Graph CNN for Learning on Point Clouds

  2018cited by this paper
• A multi-scale fully convolutional network for semantic labeling of 3D point clouds

  2018cited by this paper
• USING MULTI-SCALE FEATURES FOR THE 3D SEMANTIC LABELING OFAIRBORNE LASER SCANNING DATA

  2017cited by this paper
• A Convolutional Neural Network-Based 3D Semantic Labeling Method for ALS Point Clouds

  2017cited by this paper
• Unstructured Point Cloud Semantic Labeling Using Deep Segmentation Networks

  2017cited by this paper
• PointNet++: Deep Hierarchical Feature Learning on Point Sets in a Metric Space

  2017cited by this paper
• FAST SEMANTIC SEGMENTATION OF 3D POINT CLOUDS WITH STRONGLY VARYING DENSITY

  2016cited by this paper
• Multi-view 3D Object Detection Network for Autonomous Driving

  2016cited by this paper
• Paying More Attention to Attention: Improving the Performance of Convolutional Neural Networks via Attention Transfer

  2016cited by this paper
• OctNet: Learning Deep 3D Representations at High Resolutions

  2016cited by this paper
• Densely Connected Convolutional Networks

  2016cited by this paper
• PointNet: Deep Learning on Point Sets for 3D Classification and Segmentation

  2016cited by this paper
• HIERARCHICAL HIGHER ORDER CRF FOR THE CLASSIFICATION OF AIRBORNE LIDAR POINT CLOUDS IN URBAN AREAS

  2016cited by this paper
• VoxNet: A 3D Convolutional Neural Network for real-time object recognition

  2015cited by this paper
• Wetland inundation mapping and change monitoring using Landsat and airborne LiDAR data

  2014cited by this paper
• Contextual classification of lidar data and building object detection in urban areas

  2014influential reference
• Computationally Efficient Method for the Generation of a Digital Terrain Model From Airborne LiDAR Data Using Connected Operators

  2014cited by this paper
• Visualizing and Understanding Convolutional Networks

  2013cited by this paper
• Visual Features - From Early Concepts to Modern Computer Vision

  2013cited by this paper
• Heihe Watershed Allied Telemetry Experimental Research (HiWATER): Scientific Objectives and Experimental Design

  2013influential reference
• Fast Point Feature Histograms (FPFH) for 3D registration

  2009cited by this paper
• Tree Detection in Urban Regions Using Aerial Lidar and Image Data

  2007cited by this paper

• Optimizing Local–Global Fused Feature Space With Category-Center-Aware Supervised Contrastive Learning for Urban 3-D Mesh Segmentation

  2026cites this paper
• Multi-scale sparse convolution and point convolution adaptive fusion point cloud semantic segmentation method

  2025cites this paper
• A Semantic Segmentation Network for ALS Point Clouds Considering Geographic Location and Feature Interaction

  2025cites this paper
• HFA-Net: hybrid feature-aware network for large-scale point cloud semantic segmentation

  2025cites this paper
• APTNet: Attention-refined progressive three-branch network for real-time semantic segmentation

  2025cites this paper
• A Dual-Branch Network Leveraging Heterogeneous Features for Semantic Segmentation of Large-Scale Point Clouds

  2025cites this paper
• Local Contextual Attention for Enhancing Kernel Point Convolution in 3D Point Cloud Semantic Segmentation

  2025cites this paper