The core of self-supervised point cloud learning lies in setting up appropriate pretext tasks, to construct a pre-training framework that enables the encoder to perceive 3D objects effectively. In this paper, we integrate two prevalent methods, masked point modeling (MPM) and 3D-to-2D generation, as pretext tasks within a pre-training framework. We leverage the spatial awareness and precise supervision offered by these two methods to address their respective limitations: ambiguous supervision signals and insensitivity to geometric information. Specifically, the proposed framework, abbreviated as PointCG, consists of a Hidden Point Completion (HPC) module and an Arbitrary-view Image Generation (AIG) module. We first capture visible points from arbitrary views as inputs by removing hidden points. Then, HPC extracts representations of the inputs with an encoder and completes the entire shape with a decoder, while AIG is used to generate rendered images based on the visible points’ representations. Extensive experiments demonstrate the superiority of the proposed method over the baselines in various downstream tasks. Our code will be made available upon acceptance.@en

The conventional wisdom in point cloud analysis predominantly explores 3D geometries. It is often achieved through the introduction of intricate learnable geometric extractors in the encoder or by deepening networks with repeated blocks. However, these methods contain a significant number of learnable parameters, resulting in substantial computational costs and imposing memory burdens on CPU/GPU. Moreover, they are primarily tailored for object-level point cloud classification and segmentation tasks, with limited extensions to crucial scene-level applications, such as autonomous driving. To this end, we introduce PointeNet, an efficient network designed specifically for point cloud analysis. PointeNet distinguishes itself with its lightweight architecture, low training cost, and plug-and-play capability, while also effectively capturing representative features. The network consists of a Multivariate Geometric Encoding (MGE) module and an optional Distance-aware Semantic Enhancement (DSE) module. MGE employs operations of sampling, grouping, pooling, and multivariate geometric aggregation to lightweightly capture and adaptively aggregate multivariate geometric features, providing a comprehensive depiction of 3D geometries. DSE, designed for real-world autonomous driving scenarios, enhances the semantic perception of point clouds, particularly for distant points. Our method demonstrates flexibility by seamlessly integrating with a classification/segmentation head or embedding into off-the-shelf 3D object detection networks, achieving notable performance improvements at a minimal cost. Extensive experiments on object-level datasets, including ModelNet40, ScanObjectNN, ShapeNetPart, and the scene-level dataset KITTI, demonstrate the superior performance of PointeNet over state-of-the-art methods in point cloud analysis. Notably, PointeNet outperforms PointMLP with significantly fewer parameters on ModelNet40, ScanObjectNN, and ShapeNetPart, and achieves a substantial improvement of over 2% in 3DAP_R40 for PointRCNN on KITTI with a minimal parameter cost of 1.4 million. Code is publicly available at https://github.com/lipeng-gu/PointeNet

@en

Accurate lane maps with semantics are crucial for various applications, such as high-definition maps (HD Maps), intelligent transportation systems (ITS), and digital twins. Manual annotation of lanes is labor-intensive and costly, prompting researchers to explore automatic lane extraction methods. This paper presents an end-to-end large-scale lane mapping method that considers both lane geometry and semantics. This study represents lane markings as polylines with uniformly sampled points and associated semantics, allowing for adaptation to varying lane shapes. Additionally, we propose an end-to-end network to extract lane polylines from mobile laser scanning (MLS) data, enabling the inference of vectorized lane instances without complex post-processing. The network consists of three components: a feature encoder, a column proposal generator, and a lane information decoder. The feature encoder encodes textual and structural information of lane markings to enhance the method’s robustness to data imperfections, such as varying lane intensity, uneven point density, and occlusion-induced incomplete data. The column proposal generator generates regions of interest for the subsequent decoder. Leveraging the embedded multi-scale features from the feature encoder, the lane decoder effectively predicts lane polylines and their associated semantics without requiring step-by-step conditional inference. Comprehensive experiments conducted on three lane datasets have demonstrated the performance of the proposed method, even in the presence of incomplete data and complex lane topology. Furthermore, the datasets used in this work, including source ground points, generated bird’s eye view (BEV) images, and annotations, will be publicly available with the publication of the paper.@en

Multi-sensor object detection is an active research topic in automated driving, but the robustness of such detection models against missing sensor input (modality missing), e.g., due to a sudden sensor failure, is a critical problem which remains under-studied. In this work, we propose UniBEV, an end-to-end multi-modal 3D object detection framework designed for robustness against missing modalities: UniBEV can operate on LiDAR plus camera input, but also on LiDAR-only or camera-only input without retraining. To facilitate its detector head to handle different input combinations, UniBEV aims to create well-aligned Bird’s Eye View (BEV) feature maps from each available modality. Unlike prior BEV-based multi-modal detection methods, all sensor modalities follow a uniform approach to resample features from the original sensor coordinate systems to the BEV features. We furthermore investigate the robustness of various fusion strategies w.r.t. missing modalities: the commonly used feature concatenation, but also channel-wise averaging, and a generalization to weighted averaging termed Channel Normalized Weights. To validate its effectiveness, we compare UniBEV to state-of-the-art BEVFusion and MetaBEV on nuScenes over all sensor input combinations. In this setting, UniBEV achieves better performance than these baselines for all input combinations. An ablation study shows the robustness benefits of fusing by weighted averaging over regular concatenation, and of sharing queries between the BEV encoders of each modality. Our code is available at https://github.com/tudelft-iv/UniBEV.@en

3D modeling of indoor spaces is a prerequisite for daylight simulation, and the accuracy of the 3D models has a significant impact on the simulation. The goal of this study was to quantify the errors caused by modeling indoor spaces at different accuracy levels to find the optimal balance between the reliability of the results and labor investment. For this purpose, we introduce a level of detail (LOD) concept for indoor spaces based on the size of non-permanent indoor objects by inclusion and exclusion from the simulation scene. The errors corresponding to models with low accuracies are measured by climate-based simulation using an improved two-phase method. Our results show that inaccurate modeling of indoor spaces causes between 10-70% error in TAI with 25% median across all spaces.@en

We present PolyGNN, a polyhedron-based graph neural network for 3D building reconstruction from point clouds. PolyGNN learns to assemble primitives obtained by polyhedral decomposition via graph node classification, achieving a watertight and compact reconstruction. To effectively represent arbitrary-shaped polyhedra in the neural network, we propose a skeleton-based sampling strategy to generate polyhedron-wise queries. These queries are then incorporated with inter-polyhedron adjacency to enhance the classification. PolyGNN is end-to-end optimizable and is designed to accommodate variable-size input points, polyhedra, and queries with an index-driven batching technique. To address the abstraction gap between existing city-building models and the underlying instances, and provide a fair evaluation of the proposed method, we develop our method on a large-scale synthetic dataset with well-defined ground truths of polyhedral labels. We further conduct a transferability analysis across cities and on real-world point clouds. Both qualitative and quantitative results demonstrate the effectiveness of our method, particularly its efficiency for large-scale reconstructions.@en

Current point cloud denoising (PCD) models optimize single networks, trying to make their parameters adaptive to each point in a large pool of point clouds. Such a denoising network paradigm neglects that different points are often corrupted by different levels of noise and they may convey different geometric structures. Thus, the intricacy of both noise and geometry poses side effects including remnant noise, wrongly-smoothed edges, and distorted shape after denoising. We propose PathNet, a pathselective PCD paradigm based on reinforcement learning (RL). Unlike existing efforts, PathNet enables dynamic selection of the most appropriate denoising path for each point, best moving it onto its underlying surface. We have two more contributions besides the proposed framework of path-selective PCD for the first time. First, to leverage geometry expertise and benefit from training data, we propose a noise- and geometry-aware reward function to train the routing agent in RL. Second, the routing agent and the denoising network are trained jointly to avoid under- and over-smoothing. Extensive experiments show promising improvements of PathNet over its competitors, in terms of the effectiveness for removing different levels of noise and preserving multi-scale surface geometries. Furthermore, PathNet generalizes itself more smoothly to real scans than cutting-edge models.@en

We propose a concept of hybrid geometry sets for registering cross-source geometric data. Specifically, our method focuses on the coarse registration of geometric data obtained from laser scanning and photogrammetric reconstruction. Due to different characteristics (e.g., variations in noise levels, density, and scales), achieving accurate registration between these data becomes a challenging task. The proposed method uses geometric structures to construct hybrid geometry sets, and the geometric relations between the elements of a hybrid geometry set are encoded in a hybrid feature space. This enables effective and efficient similarity query and correspondence establishment between the hybrid geometry sets. The proposed global registration method works in three steps. Firstly, a set of hybrid geometry sets is constructed using extracted planes and intersection lines. Then the features of the hybrid geometry sets are computed to encode the relative pose and topological relationships between the extracted planes and intersection lines, and their correspondences between the two inputs are established by querying hybrid geometry sets with similar features. Finally, the global registration parameters are calculated using the correspondences, and the registration result is further refined through continuous optimization. The robustness of the method has been evaluated using different real-world cross-source geometric data of urban scenes. Extensive comparisons with state-of-the-art algorithms have also demonstrated its effectiveness.

@en

Optimizing the built environment via simulations of building models hinges on standardizing data acquisition. In this research, we put forward distinct levels of detail for geometry and material inputs, specifically tailored for indoor daylight applications. We primarily focus on understanding the uncertainties arising from imprecise estimations of material optical properties and incomplete geometrical inputs in climate-based indoor daylight simulations. Employing a Monte Carlo approach, we analyzed six office and teaching spaces, creating 20 variations for each by altering geometrical completeness and material accuracy. The technique of excluding non-permanent objects below certain sizes in four graduated steps was used to derive and test the impact of various geometrical levels of detail. Our findings reveal that different levels of geometrical completeness lead to errors ranging from 1.08% to 18.05%. Additionally, a twofold increase in simulation time was noted when geometrical detail was enhanced relative to the most basic model. Errors stemming from imprecise definitions of material optical properties showed a normal distribution. The uncertainty in simulation outcomes showed a linear rise with increasing input material uncertainty, lying between 10% to 30%, depending on space configurations. We observed heightened uncertainty near openings, attributed to window transmittance effects. The research underscores that daylight predictions are markedly more sensitive to transmittance uncertainties than to those in reflectance, regardless of the window-to-floor ratio. These insights may help to guide a more efficient data acquisition process of indoor spaces for daylight simulations.

@en

We introduce MuVieCAST, a modular multi-view consistent style transfer network architecture that enables consistent style transfer between multiple viewpoints of the same scene. This network architecture supports both sparse and dense views, making it versatile enough to handle a wide range of multi-view image datasets. The approach consists of three modules that perform specific tasks related to style transfer, namely content preservation, image transformation, and multi-view consistency enforcement. We extensively evaluate our approach across multiple application domains including depth-map-based point cloud fusion, mesh reconstruction, and novel-view synthesis. Our experiments reveal that the proposed framework achieves an exceptional generation of stylized images, exhibiting consistent outcomes across perspectives. A user study focusing on novel-view synthesis further confirms these results, with approximately 68% of cases participants expressing a preference for our generated outputs compared to the recent state-of-the-art method. Our modular framework is extensible and can easily be integrated with various backbone architectures, making it a flexible solution for multi-view style transfer.@en

The acquisition of both local and global features from irregular point clouds is crucial for 3D object detection (3DOD). Current mainstream 3D detectors neglect significant local features during pooling operations or disregard many global features of the overall scene context. This paper proposes new techniques for simultaneously learning local-global features of scene point clouds to enhance 3DOD. Specifically, we propose an efficient 3DOD network in indoor point clouds, named SimLOG, which utilizes simultaneous local-global feature learning. SimLOG has two main contributions: a Dynamic Points Interaction (DPI) module to recover local features lost during pooling, and a Global Context Aggregation(GCA) module to aggregate multi-scale features from various layers of the encoder to improve scene context awareness. Unlike traditional local-global feature learning methods, our DPI and GCA modules are integrated into a single feature learning module, making it easily detachable and able to be incorporated into existing 3DOD networks to enhance their performance. SimLOG demonstrates superior performance over twenty competitors in terms of detection accuracy and robustness on both the SUN RGB-D and ScanNet V2 datasets. Specifically, SimLOG boosts the baseline VoteNet by 8.1% of mAP@0.25 on ScanNet V2 and by 3.9% of mAP@0.25 on SUN RGB-D.@en

We propose an enhancement module called depth discontinuity learning (DDL) for learning-based multi-view stereo (MVS) methods. Traditional methods are known for their accuracy but struggle with completeness. While recent learning-based methods have improved completeness at the cost of accuracy, our DDL approach aims to improve accuracy while retaining completeness in the reconstruction process. To achieve this, we introduce the joint estimation of depth and boundary maps, where the boundary maps are explicitly utilized for further refinement of the depth maps. We validate our idea by integrating it into an existing learning-based MVS pipeline where the reconstruction depends on high-quality depth map estimation. Extensive experiments on various datasets, namely DTU, ETH3D, “Tanks and Temples”, and BlendedMVS, show that our method improves reconstruction quality compared to our baseline, Patchmatchnet. Our ablation study demonstrates that incorporating the proposed DDL significantly reduces the depth map error, for instance, by more than 30% on the DTU dataset, and leads to improved depth map quality in both smooth and boundary regions. Additionally, our qualitative analysis has shown that the reconstructed point cloud exhibits enhanced quality without any significant compromise on completeness. Finally, the experiments reveal that our proposed model and strategies exhibit strong generalization capabilities across the various datasets.@en

Registering point clouds of forest environments is an essential prerequisite for LiDAR applications in precision forestry. State-of-the-art methods for forest point cloud registration require the extraction of individual tree attributes, and they have an efficiency bottleneck when dealing with point clouds of real-world forests with dense trees. We propose an automatic, robust, and efficient method for the registration of forest point clouds. Our approach first locates tree stems from raw point clouds and then matches the stems based on their relative spatial relationship to determine the registration transformation. The algorithm requires no extra individual tree attributes and has quadratic complexity to the number of trees in the environment, allowing it to align point clouds of large forest environments. Extensive experiments on forest terrestrial point clouds have revealed that our method inherits the effectiveness and robustness of the stem-based registration strategy while exceedingly increasing its efficiency. Besides, we introduce a new benchmark dataset that complements the very few existing open datasets for the development and evaluation of registration methods for forest point clouds. The source code of our method and the dataset are available at https://github.com/zexinyang/GlobalMatch.

@en

We introduce a novel deep learning-based framework to interpret 3D urban scenes represented as textured meshes. Based on the observation that object boundaries typically align with the boundaries of planar regions, our framework achieves semantic segmentation in two steps: planarity-sensible over-segmentation followed by semantic classification. The over-segmentation step generates an initial set of mesh segments that capture the planar and non-planar regions of urban scenes. In the subsequent classification step, we construct a graph that encodes the geometric and photometric features of the segments in its nodes and the multi-scale contextual features in its edges. The final semantic segmentation is obtained by classifying the segments using a graph convolutional network. Experiments and comparisons on two semantic urban mesh benchmarks demonstrate that our approach outperforms the state-of-the-art methods in terms of boundary quality, mean IoU (intersection over union), and generalization ability. We also introduce several new metrics for evaluating mesh over-segmentation methods dedicated to semantic segmentation, and our proposed over-segmentation approach outperforms state-of-the-art methods on all metrics. Our source code is available at https://github.com/WeixiaoGao/PSSNet.

@en

How will you repair a physical object with some missings? You may imagine its original shape from previously captured images, recover its overall (global) but coarse shape first, and then refine its local details. We are motivated to imitate the physical repair procedure to address point cloud completion. To this end, we propose a cross-modal shape-transfer dual-refinement network (termed CSDN), a coarse-to-fine paradigm with images of full-cycle participation, for quality point cloud completion. CSDN mainly consists of “shape fusion” and “dual-refinement” modules to tackle the cross-modal challenge. The first module transfers the intrinsic shape characteristics from single images to guide the geometry generation of the missing regions of point clouds, in which we propose IPAdaIN to embed the global features of both the image and the partial point cloud into completion. The second module refines the coarse output by adjusting the positions of the generated points, where the local refinement unit exploits the geometric relation between the novel and the input points by graph convolution, and the global constraint unit utilizes the input image to fine-tune the generated offset. Different from most existing approaches, CSDN not only explores the complementary information from images but also effectively exploits cross-modal data in the whole coarse-to-fine completion procedure. Experimental results indicate that CSDN performs favorably against twelve competitors on the cross-modal benchmark.@en

Visual place recognition (VPR) is an image-based localization method that estimates the camera location of a query image by retrieving the most similar reference image from a map of geo-tagged reference images. In this work, we look into two fundamental bottlenecks for its localization accuracy: 1) reference map sparseness and 2) viewpoint invariance. First, the reference images for VPR are only available at sparse poses in a map, which enforces an upper bound on the maximum achievable localization accuracy through VPR. We, therefore, propose Continuous Place-descriptor Regression (CoPR) to densify the map and improve localization accuracy. We study various interpolation and extrapolation models to regress additional VPR feature descriptors from only the existing references. Second, we compare different feature encoders and show that CoPR presents value for all of them. We evaluate our models on three existing public datasets and report on average around 30% improvement in VPR-based localization accuracy using CoPR, on top of the 15% increase by using a viewpoint-variant loss for the feature encoder. The complementary relation between CoPR and relative pose estimation is also discussed.@en

Continuous implicit representations can flexibly describe complex 3D geometry and offer excellent potential for 3D point cloud analysis. However, it remains challenging for existing point-based deep learning architectures to leverage the implicit representations due to the discrepancy in data structures between implicit fields and point clouds. In this work, we propose a new point cloud representation by integrating the 3D Cartesian coordinates with the intrinsic geometric information encapsulated in its implicit field. Specifically, we parameterize the continuous unsigned distance field around each point into a low-dimensional feature vector that captures the local geometry. Then we concatenate the 3D Cartesian coordinates of each point with its encoded implicit feature vector as the network input. The proposed method can be plugged into an existing network architecture as a module without trainable weights. We also introduce a novel local canonicalization approach to ensure the transformation-invariance of encoded implicit features. With its local mechanism, our implicit feature encoding module can be applied to not only point clouds of single objects but also those of complex real-world scenes. We have validated the effectiveness of our approach using five well-known point-based deep networks (i.e., PointNet, SuperPoint Graph, RandLA-Net, CurveNet, and Point Structuring Net) on object-level classification and scene-level semantic segmentation tasks. Extensive experiments on both synthetic and real-world datasets have demonstrated the effectiveness of the proposed point representation.

@en

Symmetry widely exists in nature and man-made shapes, but it is unavoidably distorted during the process of growth, design, digitalization, and reconstruction steps. To enhance symmetry, traditional methods follow the detect-then-symmetrize paradigm, which is sensitive to noise in the detection phase, resulting in ambiguities for the subsequent symmetrization step. In this work, we propose a novel optimization-based framework that jointly detects and optimizes symmetry for 2D shapes represented as polygons. Our method can detect and optimize symmetry using a single objective function. Specifically, we formulate symmetry detection and optimization as a mixed-integer program. Our method first generates a set of candidate symmetric edge pairs, which are then encoded as binary variables in our optimization. The geometry of the shape is expressed as continuous variables, which are then optimized together with the binary variables. The symmetry of the shape is enforced by the designed hard constraints. After the optimization, both the optimal symmetric edge correspondences and the geometry are obtained. Our method simultaneously detects all the symmetric primitive pairs and enhances the symmetry of a model while minimally altering its geometry. We have tested our method on a variety of shapes from designs and vectorizations, and the results have demonstrated its effectiveness.

@en

Complexity of forest structure is an important factor contributing to uncertainty in aboveground biomass estimates. In this study, we present a new method for reducing uncertainty in forest aboveground biomass (AGB) estimation based on plot-level terrestrial laser scanner (TLS) point clouds reconstruction. The method estimates the total AGB of plots with complex structures after automatically performing the steps of ground point filtering, single tree segmentation, and three-dimensional (3D) structure reconstruction. We used plot data from temperate and tropical forest ecosystems to verify the effectiveness of the method, reconstructing a 1300 m² temperate plantation plot and a 5000 m² mingled forest plot, respectively. The total biomass of 153 trees in the plantation plot was overestimated by 17.12 %, and the total biomass of 61 trees in the mingled forest plot was underestimated by 10.88 %. We found that the uncertainty of aboveground biomass estimation in tropical forests with more complex structures is not necessarily greater than in plantations. Therefore, in large-scale remote sensing observations of forest biomass, the number or area of plots can be increased to reduce the uncertainty of the results caused by the complex structure. The focus of this study is to explore TLS point clouds modeling methods to reduce the uncertainty in AGB estimation caused by the complexity of forest structures, and to provide reference cases for plot-level point clouds reconstruction methods. Forest ecologists can use this method to regularly observe forest growth and obtain indicators related to forest ecology without destroying trees.

@en

This paper presents a method for multiple object tracking (MOT) in video streams. The method incorporates the prediction of physical locations of people into a tracking-by-detection paradigm. We predict the trajectories of people on an estimated ground plane and apply a learning-based network to extract the appearance features across frames. The method transforms the detected object locations from image space to an estimated ground space to refine the tracking trajectories. This transform space allows the objects detected from multi-view images to be associated under one coordinate system. Besides, the occluded pedestrians in image space can be well separated in a rectified ground plane where the motion models of the pedestrians are estimated. The effectiveness of this method is evaluated on different datasets by extensive comparisons with state-of-The-Art techniques. Experimental results show that the proposed method improves MOT tasks in terms of the number of identity switches (IDSW) and the fragmentations (Frag).

@en