Linking Points With Labels in 3D: A Review of Point Cloud Semantic Segmentation

Abstract

3D Point Cloud Semantic Segmentation (PCSS) is attracting increasing interest, due to its applicability in remote sensing, computer vision and robotics, and due to the new possibilities offered by deep learning techniques. In order to provide a needed up-to-date review of recent developments in PCSS, this article summarizes existing studies on this topic. Firstly, we outline the acquisition and evolution of the 3D point cloud from the perspective of remote sensing and computer vision, as well as the published benchmarks for PCSS studies. Then, traditional and advanced techniques used for Point Cloud Segmentation (PCS) and PCSS are reviewed and compared. Finally, important issues and open questions in PCSS studies are discussed.

arXiv commentThe title of published version was modified to "Linking Points With Labels in 3D: A Review of Point Cloud Semantic Segmentation". To read its final version please go to IEEE Geoscience and Remote Sensing Magazine on IEEE XPlore: https://ieeexplore.ieee.org/document/9028090

References in corpus (5)

• Deep learning in remote sensing: a review
• O-CNN: Octree-based Convolutional Neural Networks for 3D Shape Analysis
• Exploring Spatial Context for 3D Semantic Segmentation of Point Clouds
• Supervised Fitting of Geometric Primitives to 3D Point Clouds
• Non-Local Compressive Sensing Based SAR Tomography

Cited by in corpus (25)

• Unsupervised Point Cloud Representation Learning with Deep Neural Networks: A Survey
• The Hessigheim 3D (H3D) Benchmark on Semantic Segmentation of High-Resolution 3D Point Clouds and Textured Meshes from UAV LiDAR and Multi-View-Stereo
• A comprehensive overview of deep learning techniques for 3D point cloud classification and semantic segmentation
• Deep Learning for 3D Point Clouds: A Survey
• Deep Learning Based 3D Segmentation: A Survey
• A Survey on Occupancy Perception for Autonomous Driving: The Information Fusion Perspective
• Clustering performance analysis using a new correlation-based cluster validity index
• Deep Learning on 3D Semantic Segmentation: A Detailed Review
• Tinto: Multisensor Benchmark for 3D Hyperspectral Point Cloud Segmentation in the Geosciences
• Juggling With Representations: On the Information Transfer Between Imagery, Point Clouds, and Meshes for Multi-Modal Semantics
• Variational Shape Approximation of Point Set Surfaces
• SnapshotNet: Self-supervised Feature Learning for Point Cloud Data Segmentation Using Minimal Labeled Data
• A Deep Learning Approach to Identify Rock Bolts in Complex 3D Point Clouds of Underground Mines Captured Using Mobile Laser Scanners
• From CAD models to soft point cloud labels: An automatic annotation pipeline for cheaply supervised 3D semantic segmentation
• Irregular Change Detection in Sparse Bi-Temporal Point Clouds using Learned Place Recognition Descriptors and Point-to-Voxel Comparison
• A Bayesian cluster validity index
• Are We Hungry for 3D LiDAR Data for Semantic Segmentation? A Survey and Experimental Study
• Automatic Segmentation of Aircraft Dents in Point Clouds
• Self-Supervised Pre-Training Boosts Semantic Scene Segmentation on LiDAR Data
• LiDAR Remote Sensing Meets Weak Supervision: Concepts, Methods, and Perspectives
• DSOR: A Scalable Statistical Filter for Removing Falling Snow from LiDAR Point Clouds in Severe Winter Weather
• A correlation-based fuzzy cluster validity index with secondary options detector
• A Survey on Deep Learning Architectures for Point Cloud Classification and Segmentation
• PointManifoldCut: Point-wise Augmentation in the Manifold for Point Clouds
• HINORA, a method for detecting ring-like structures in 3D point distributions I: application to the Local Volume Galaxy catalogue