Multi-Modal LiDAR Point Cloud Semantic Segmentation with Salience Refinement and Boundary Perception

Point cloud has emerged as the most popular three-dimensional (3D) data format in recent years for several scientific and industrial applications. Point cloud semantic segmentation has piqued the researcher’s interest, which is a crucial stage in 3D analysis and scene comprehension. Deep learning-based processing is more feasible to increase the availability of point cloud acquisition tools that is LiDAR systems at the user end. The point cloud learning achieves tremendous success in object detection, object categorization, and semantic segmentation. To summarize the recent works with chronological development, comprehensive review of projection-, voxel-, and direct point-based point cloud semantic segmentation methods is performed from various perspectives. The commonly used point cloud benchmark datasets with their characteristics are discussed, and they are used for the performance analysis and comparison of several state-of-the-art segmentation methods. The quantitative performance analysis of these deep learning models summarizes the trend of semantic segmentation of point clouds. In the context of point cloud semantic segmentation, the various methods have specific roles. Based on the review of methods working and their performance analysis, it is concluded that the projection-based methods prioritize efficiency, which is ideal in unavailability of high-performance computing system. Voxel-based methods capture overall context, serving well in 3D object classification. Point-based approaches excel in fine details and efficiency, suited for tasks like 3D semantic segmentation. Choosing the suitable method depends on the task, data, and resources. KPConv and DGCNN are popular choices, especially for precision and adaptability to point density. However, method performance varies, underlining the need for tailored selection. Hybrid approaches, combining method strengths, promise superior results.

Three-dimensional (3D) semantic segmentation of point clouds is important in many scenarios, such as automatic driving, robotic navigation, while edge computing is indispensable in the devices. Deep learning methods based on point sampling prove to be computation and memory efficient to tackle large-scale point clouds (e.g. millions of points). However, some local features may be abandoned while sampling. In this paper, We present one end-to-end 3D semantic segmentation framework based on dilated nearest neighbor encoding. Instead of down-sampling point cloud directly, we propose a dilated nearest neighbor encoding module to broaden the network’s receptive field to learn more 3D geometric information. Without increase of network parameters, our method is computation and memory efficient for large-scale point clouds. We have evaluated the dilated nearest neighbor encoding in two different networks. The first is the random sampling with local feature aggregation. The second is the Point Transformer. We have evaluated the quality of the semantic segmentation on the benchmark 3D dataset S3DIS, and demonstrate that the proposed dilated nearest neighbor encoding exhibited stable advantages over baseline and competing methods.

Deep learning network for indoor point cloud semantic segmentation with transferability

LESA-Net: Semantic segmentation of multi-type road point clouds in complex agroforestry environment

The semantic segmentation of point clouds is a crucial undertaking in 3D reconstruction and holds great importance. However, achieving precise semantic segmentation represents a significant hurdle. In this paper, we present BEMF-Net, an efficient method for large-scale environments. It starts with an effective feature extraction method. Unlike images, 3D data comprise not only geometric relations but also texture information. To accurately depict the scene, it is crucial to take into account the impacts of texture and geometry on the task, and incorporate modifications to improve feature description. Additionally, we present a multi-scale feature fusion technique that effectively promotes the interaction between features at different resolutions. The approach mitigates the problem of the smoothing of detailed information caused by downsampling mechanisms, while ensuring the integrity of features across different layers, allowing a more comprehensive representation of the point cloud. We confirmed the effectiveness of this method by testing it on benchmark datasets such as S3DIS, SensatUrban, and Toronto3D.

Semantic segmentation of large-scale point clouds by integrating attention mechanisms and transformer models

Point cloud semantic segmentation in urban scenes plays a vital role in intelligent city modeling, autonomous driving, and urban planning. Point cloud semantic segmentation based on deep learning methods has achieved significant improvement. However, it is also challenging for accurate semantic segmentation in large scenes due to complex elements, variety of scene classes, occlusions, and noise. Besides, most methods need to split the original point cloud into multiple blocks before processing and cannot directly deal with the point clouds on a large scale. We propose a novel context-aware network (CAN) that can directly deal with large-scale point clouds. In the proposed network, a Local Feature Aggregation Module (LFAM) is designed to preserve rich geometric details in the raw point cloud and reduce the information loss during feature extraction. Then, in combination with a Global Context Aggregation Module (GCAM), capture long-range dependencies to enhance the network feature representation and suppress the noise. Finally, a Context-Aware Upsampling Module (CAUM) is embedded into the proposed network to capture the global perception from a broad perspective. The ensemble of low-level and high-level features facilitates the effectiveness and efficiency of 3D point cloud feature refinement. Comprehensive experiments were carried out on three large-scale point cloud datasets in both outdoor and indoor environments to evaluate the performance of the proposed network. The results show that the proposed method outperformed the state-of-the-art representative semantic segmentation networks, and the overall accuracy (OA) of Tongji-3D, Semantic3D, and S3DIS is 96.01%, 95.0%, and 88.55%, respectively.

Point clouds have become essential for digital heritage preservation, aiding in the identification and classification of complex structural elements. However, most datasets rely on single-modal data, limiting their ability to describe real-world scenarios comprehensively. This paper introduces the Real-World Multi-modal Ancient Architecture Point Cloud Semantic Segmentation Dataset (RW-MAPCSD), which includes multi-modal data such as point clouds, line drawings, color, and depth projections, enabling multi-modal analysis of ancient buildings. To address data imbalance, we propose a novel segmentation network, Mask2former-KNN 3D Network (MK3DNet). The network projects point clouds into images while preserving point indices, using image segmentation techniques for initial segmentation. Results are then mapped back to the point cloud and refined with the K-nearest neighbors (KNN) algorithm. Experimental results show significant improvements, with mIoU and OA of 77.47% and 90.85%, respectively, surpassing the Point Transformer network by 21.94% and 5.87%.

Road geometric information and a digital model based on light detection and ranging (LiDAR) can perform accurate geometric inventories and three-dimensional (3D) descriptions for as-built roads and infrastructures. However, unorganized point clouds and complex road scenarios would reduce the accuracy of geometric information extraction and digital modeling. There is a standardization need for information extraction and 3D model construction that integrates point cloud processing and digital modeling. This paper develops a framework from semantic segmentation to geometric information extraction and digital modeling based on LiDAR data. A semantic segmentation network is improved for the purpose of dividing the road surface and infrastructure. The road boundary and centerline are extracted by the alpha-shape and Voronoi diagram methods based on the semantic segmentation results. The road geometric information is obtained by a coordinate transformation matrix and the least square method. Subsequently, adaptive road components are constructed using Revit software. Thereafter, the road route, road entity model, and various infrastructure components are generated by the extracted geometric information through Dynamo and Revit software. Finally, a detailed digital model of the road scenario is developed. The Toronto-3D and Semantic3D datasets are utilized for analysis through training and testing. The overall accuracy (OA) of the proposed net for the two datasets is 95.3 and 95.0%, whereas the IoU of segmented road surfaces is 95.7 and 97.9%. This indicates that the proposed net could accomplish superior performance for semantic segmentation of point clouds. The mean absolute errors between the extracted and manually measured geometric information are marginal. This demonstrates the effectiveness and accuracy of the proposed extraction methods. Thus, the proposed framework could provide a reference for accurate extraction and modeling from LiDAR data.

Point cloud data provide three-dimensional (3D) information about objects in the real world, containing rich semantic features. Therefore, the task of semantic segmentation of point clouds has been widely applied in fields such as robotics and autonomous driving. Although existing research has made unprecedented progress, achieving real-time semantic segmentation of point clouds on airborne devices still faces challenges due to excessive computational and memory requirements. To address this issue, we propose a novel sequence convolution semantic segmentation architecture that integrates Convolutional Neural Networks (CNN) with a sequence-to-sequence (seq2seq) structure, termed SeqConv-Net. This architecture views point cloud semantic segmentation as a sequence generation task. Based on our unique perspective of spatially ordered sequences, we use Recurrent Neural Networks (RNN) to encode elevation information, then input the structured hidden states into a CNN for planar feature extraction. The results are combined with the RNN’s encoded outputs via residual connections and are fed into a decoder for sequence prediction in a seq2seq manner. Experiments show that the SeqConv-Net architecture achieves 75.5% mean Intersection Over Union (mIOU) accuracy on the DALES dataset, with the total processing speed from data preprocessing to prediction being several to tens of times faster than existing methods. Additionally, SeqConv-Net can balance accuracy and speed by adjusting the hyperparameters and using different RNNs and CNNs, providing a new solution for real-time point cloud semantic segmentation in airborne environments.

With the development of airborne light detection and ranging (LiDAR) technology, it has become a common and efficient way to collect large-scale 3D spatial information. However, efficient and automatic semantic segmentation of LiDAR data, in the form of 3D point clouds, remains a persistent challenge. To address this, a dual attention neural network (DA-Net) is proposed, consisting of two different blocks, namely augmented edge representation (AER) and elevation attentive pooling (EAP). First, the AER can adaptively represent local orientation and position, thereby effectively enhancing geometric information. Second, the captured local features of centroid points are utilized to further encode discriminative features using the EAP with the learned attention scores. Finally, a location homogeneity (LH) module is devised to explore the long-range relationship in an encoder-decoder network. Benefiting from the dual attention module, geometric information hidden in unorganized point clouds can be effectively propagated. Besides, the LH forces the network to pay attention to the semantic consistency of elevated objects, which facilitates both point- and object-level point cloud semantic segmentation for scene understanding. A benchmark dataset is used to assess the proposed method, which achieves an overall accuracy of 85.98% and an average F1 score of 72.31%. In addition, comparisons with other latest deep learning methods on the 2019 Data Fusion Contest dataset further demonstrate the robustness and generalization ability of the proposed method.

Accurate semantic segmentation of 3D point clouds is a long-standing problem in remote sensing and computer vision. Due to the unstructured nature of point clouds, designing deep neural architectures for point cloud semantic segmentation is often not straightforward. In this work, we circumvent this problem by devising a technique to exploit structured neural architectures for unstructured data. In particular, we employ the popular convolutional neural network (CNN) architectures to perform semantic segmentation of LiDAR data. We propose a projection-based scheme that performs an angle-wise slicing of large 3D point clouds and transforms those slices into 2D grids. Accounting for intensity and reflectivity of the LiDAR input, the 2D grid allows us to construct a pseudo image for the point cloud slice. We enhance this image with low-level image processing techniques of normalization, histogram equalization, and decorrelation stretch to suit our ultimate object of semantic segmentation. A large number of images thus generated are used to induce an encoder-decoder CNN model that learns to compute a segmented 2D projection of the scene, which we finally back project to the 3D point cloud. In addition to a novel method, this article also makes a second major contribution of introducing the enhanced version of our large-scale public PC-Urban outdoor dataset which is captured in a civic setup with an Ouster LiDAR sensor. The updated dataset (PC-Urban_V2) provides nearly 8 billion points including over 100 million points labeled for 25 classes of interest. We provide a thorough evaluation of our technique on PC-Urban_V2 and three other public datasets.

Accurate organ-level segmentation is essential for achieving high-throughput, non-destructive, and automated plant phenotyping. To address the challenge of intelligent acquisition of phenotypic parameters in tomato plants, we propose TSINet, an end-to-end dual-task segmentation network designed for effective and precise semantic labeling and instance recognition of tomato point clouds, based on the Pheno4D dataset. TSINet adopts an encoder–decoder architecture, where a shared encoder incorporates four Geometry-Aware Adaptive Feature Extraction Blocks (GAFEBs) to effectively capture local structures and geometric relationships in raw point clouds. Two parallel decoder branches are employed to independently decode shared high-level features for the respective segmentation tasks. Additionally, a Dual Attention-Based Feature Enhancement Module (DAFEM) is introduced to further enrich feature representations. The experimental results demonstrate that TSINet achieves superior performance in both semantic and instance segmentation, particularly excelling in challenging categories such as stems and large-scale instances. Specifically, TSINet achieves 97.00% mean precision, 96.17% recall, 96.57% F1-score, and 93.43% IoU in semantic segmentation and 81.54% mPrec, 81.69% mRec, 81.60% mCov, and 86.40% mWCov in instance segmentation. Compared with state-of-the-art methods, TSINet achieves balanced improvements across all metrics, significantly reducing false positives and false negatives while enhancing spatial completeness and segmentation accuracy. Furthermore, we conducted ablation studies and generalization tests to systematically validate the effectiveness of each TSINet component and the overall robustness of the model. This study provides an effective technological approach for high-throughput automated phenotyping of tomato plants, contributing to the advancement of intelligent agricultural management.

Performing semantic segmentation on point clouds is the primary method by which machines perceive 3D scenes in a fine-grained manner. Deep learning algorithms usually require many pointwise annotations obtained with specialized tools, which is a laborious and inefficient process. To this end, we develop two frameworks for training point cloud semantic segmentation networks, one that utilizes fewer projected image annotations and another that employs sparse scribble image annotations, making the process more flexible and user friendly. However, back-projecting 2D-pixel labels to 3D points during loss calculations always introduces errors. To increase the back-projection accuracy of our approach, we first identify and record potential pixel-point correspondence errors and then develop strategies for constructing an accurate back-projection mapping matrix. Specifically, we filter out occluded and noisy points to avoid incorrect label allocations and permit multiclass assignments to adjust the ambiguity of boundary points. By incorporating an accurate back-projection mechanism into the loss functions of the proposed training frameworks, our networks can perform well with only four projected image annotations or even sparse scribble image annotations for each scene. This results in state-of-the-art performance compared with that of other weakly supervised point cloud semantic segmentation approaches, and the outcomes are even comparable to those produced by fully supervised methods on the S3DIS and ScanNet-v2 datasets.

Abstract3D point cloud semantic segmentation is fundamental for autonomous driving. Most approaches in the literature neglect an important aspect, i.e., how to deal with domain shift when handling dynamic scenes. This can significantly hinder the navigation capabilities of self-driving vehicles. This paper advances the state of the art in this research field. Our first contribution consists in analysing a new unexplored scenario in point cloud segmentation, namely Source-Free Online Unsupervised Domain Adaptation (SF-OUDA). We experimentally show that state-of-the-art methods have a rather limited ability to adapt pre-trained deep network models to unseen domains in an online manner. Our second contribution is an approach that relies on adaptive self-training and geometric-feature propagation to adapt a pre-trained source model online without requiring either source data or target labels. Our third contribution is to study SF-OUDA in a challenging setup where source data is synthetic and target data is point clouds captured in the real world. We use the recent SynLiDAR dataset as a synthetic source and introduce two new synthetic (source) datasets, which can stimulate future synthetic-to-real autonomous driving research. Our experiments show the effectiveness of our segmentation approach on thousands of real-world point clouds (Code and synthetic datasets are available at https://github.com/saltoricristiano/gipso-sfouda).KeywordsOnline domain adaptationSource-free unsupervised domain adaptationPoint cloud segmentationGeometric propagation