Point cloud data have been extensively studied due to their compact form and flexibility in representing complex 3D geometries and structures. The ability of point cloud data to accurately capture and represent intricate 3D geometry makes it an ideal choice for a wide range of applications, including 3D computer graphics, autonomous driving, robotics, and augmented reality, all of which require an understanding of the underlying geometry and spatial structures. Given the challenges associated with annotating large-scale point clouds, self-supervised point cloud representation learning has attracted increasing attention in recent years. It aims to learn generic and useful point cloud representations from unlabeled data, circumventing the need for extensive manual annotation. In this paper, we present a comprehensive survey of self-supervised point cloud representation learning using DNNs. We begin by presenting the motivation and general trends in recent research, then briefly introduce commonly used datasets and evaluation metrics. Next, we extensively explore self supervised point cloud representation learning methods. Finally, we share our thoughts on some of the challenges and potential issues that future research into self supervised learning for pre-training 3D point clouds may encounter. Our curated bibliography can be found at https://github.com/EtronTech/Awesome_3DSSL.

Noise is an inevitable aspect of point cloud acquisition, necessitating filtering as a fundamental task within the realm of 3D vision. Existing learning-based filtering methods have shown promising capabilities on commonly used datasets. Nonetheless, the effectiveness of these methods is constrained when dealing with a substantial quantity of point clouds. This limitation primarily stems from their limited denoising capabilities for dense and large-scale point clouds and their inclination to generate noisy outliers after denoising. To deal with this challenge, we introduce 3DMambaIPF, for the first time, exploiting Selective State Space Models (SSMs) architecture to handle highly-dense and large-scale point clouds, capitalizing on its strengths in selective input processing and large context modeling capabilities. Additionally, we present a robust and fast differentiable rendering loss to constrain the noisy points around the surface. In contrast to previous methodologies, this differentiable rendering loss enhances the visual realism of denoised geometric structures and aligns point cloud boundaries more closely with those observed in real-world objects. Extensive evaluations on commonly used datasets (typically with up to 50K points) demonstrate that 3DMambaIPF achieves state-of-the-art results. Moreover, we showcase the superior scalability and efficiency of 3DMambaIPF on highly dense and large-scale point clouds with up to 500K points compared to off-the-shelf methods.

Novel view synthesis (NVS) and surface reconstruction (SR) are essential tasks in 3D Gaussian Splatting (3D-GS). Despite recent progress, these tasks are often addressed independently, with GS-based rendering methods struggling under diverse light conditions and failing to produce accurate surfaces, while GS-based reconstruction methods frequently compromise rendering quality. This raises a central question: must rendering and reconstruction always involve a trade-off? To address this, we propose MGSR, a 2D/3D Mutual-boosted Gaussian splatting for Surface Reconstruction that enhances both rendering quality and 3D reconstruction accuracy. MGSR introduces two branches--one based on 2D-GS and the other on 3D-GS. The 2D-GS branch excels in surface reconstruction, providing precise geometry information to the 3D-GS branch. Leveraging this geometry, the 3D-GS branch employs a geometry-guided illumination decomposition module that captures reflected and transmitted components, enabling realistic rendering under varied light conditions. Using the transmitted component as supervision, the 2D-GS branch also achieves high-fidelity surface reconstruction. Throughout the optimization process, the 2D-GS and 3D-GS branches undergo alternating optimization, providing mutual supervision. Prior to this, each branch completes an independent warm-up phase, with an early stopping strategy implemented to reduce computational costs. We evaluate MGSR on a diverse set of synthetic and real-world datasets, at both object and scene levels, demonstrating strong performance in rendering and surface reconstruction. Code is available at https://github.com/TsingyuanChou/MGSR.

Unsupervised domain adaptation (UDA) is vital for alleviating the workload of labeling 3D point cloud data and mitigating the absence of labels when facing an unseen domain. Various methods have recently emerged to utilize images along with point clouds to enhance the performance of cross-domain 3D segmentation. However, the pseudo labels, which are generated from models trained on the source domain and provide additional supervised signals for the target domain, are inadequate when utilized for 3D segmentation due to their inherent noisiness and consequently restrict the accuracy of neural networks. With the advent of 2D Visual Foundation Models (VFMs) and their abundant knowledge prior, we propose a novel pipeline VFMSeg to further enhance the cross-modal UDA framework by leveraging these models. In this work, we study how to harness the knowledge priors learned by VFMs to produce more accurate labels for unlabeled target domains and improve overall performance. We first utilize a multi-modal VFM, which is pre-trained on large-scale image-text pairs, to provide supervised labels (VFM-PL) for images and point clouds from the target domain. Then, we adopt another VFM to generate fine-grained 2D masks for guiding the generation of augmented images and point clouds, which mix the data from source and target domains like view frustums (FrustumMixing). Finally, we merge class-wise prediction across modalities to produce more accurate annotations for unlabeled target domains. Our method is evaluated on various autonomous driving datasets and the results demonstrate a significant improvement in 3D segmentation task. Our code is available athttps://github.com/EtronTech/VFMSeg

3D Gaussian Splatting (3D-GS) has emerged as a significant advancement in the field of computer graphics and 3D vision, offering explicit scene representation and novel view synthesis without the reliance on neural networks. This technique has found diverse applications in areas such as robotics, urban mapping, autonomous navigation, and virtual reality/augmented reality, just name a few. Given the growing popularity and expanding research in 3D-GS, this paper presents a comprehensive survey of relevant papers from the past year. We organize the survey into taxonomies based on characteristics and applications, providing an introduction to the theoretical underpinnings of 3D-GS. The survey aims to introduce the theoretical foundations of 3D Gaussian Splatting and provide a reference for new researchers while inspiring future research directions.

This special issue focuses on all aspects on techniques and applications in cyber intelligence research. The purpose of this special issue is to provide a forum for presentation and discussion of innovative ideas, cutting-edge research results, and novel techniques, methods, and applications on all aspects of cyber intelligence. It includes selected papers from the 2018 International Conference on Applications and Techniques in Cyber Intelligence (ATCI2018) July 11-13, 2018, Shanghai, China, as well as an open call. The submitted manuscripts were reviewed by experts from both academia and industry. After two rounds of reviewing, the ten highest quality manuscripts were suggested for the editor in chief to accept for this special issue. This special issue will be published by Concurrency and Computation: Practice and Experience (CCPE) as special issues. A kind of cutting and nesting system for cloud manufacturing based on service-oriented architecture (SOA) is proposed by Li and Wang1, which deploys a nesting algorithm distributedly to a multicore computer cluster to provide the cutting and nesting service for numerous small and medium enterprises through uniform network service and to realize the integration with an existing enterprise system based on SOA. Wang and Zhu2 propose an improved term frequency-inverse document frequency (TFIDF) model combined with the Word2vec model for weighing word vectors. In view of the inability of the Word2vec model to distinguish the importance of words with the text, TFIDF is further introduced to weighing Word2vec word vectors to achieve a weighted Word2vec classification model. To ensure the navigational safety of ships in the sea and reduce human errors, a set of navigational information systems with decision support function is developed by Liu et al.3 A supervised learning framework with strong expansibility is established by Jia and Yang4 for search-engine joint ranking problem. It can transform existing algorithms into corresponding learning algorithms, and design new algorithms under this framework. Jiang et al5 present the game model in the social network environment by analyzing the decision-making process of conflict events and the factors to describe the evolution of conflict events. The interaction design of a driver communication system is studied by Zhao and Lin.6 They carried out three driving simulation experiments to study three different ways of interaction interfaces and then promoted a questionnaire based on NASA Text Load Index to understand the subjective feelings and subjective workload of drivers in using these interaction interfaces. Xiaoyan7 does the research on the mechanism of government subsidies and carbon tax influence on producers' decision-making by establishing an evolutionary game model between governments and producers and by integrating with prospect theory. Hu and Liu8 propose an E-maintenance framework for public infrastructure, combining Building Information Modeling, Industry Foundation Classes, and Semantic Web technologies to help integrate heterogeneous data and expert knowledge, enable information sharing through the whole life cycle, and support maintenance managers to make effective maintenance decisions. Darabian et al9 use sequential pattern mining technique to detect most frequent opcode sequences of malicious Internet of things applications. Qiu et al10 prove that the local diagnosability of a matching composition network can be obtained by adding one to that of the component with some structural restrictions. The guest editors would like to thank Professor Geoffrey C. Fox who is the editor in chief of CCPE. His help and trust are the most important thing for the success of this special issue. The guest editors would like to thank the reviewers for their high-quality reviews, which provided insightful and constructive feedback to the authors of the papers.

This special issue focuses on all aspects on techniques and applications in cyber intelligence research. The purpose of this special issue is to provide a forum for presentation and discussion of innovative ideas, cutting edge research results, and novel techniques, methods and applications on all aspects of cyber intelligence. It includes selected articles from 2019 International Conference on Applications and Techniques in Cyber Intelligence (ATCI2019) June 22-24, 2019 Huainan, China as well as an open call. The submitted articles were reviewed by experts from both academia and industry. After two rounds of reviewing, the 15 highest quality manuscripts were suggested for EiC to accept for this special issue. This special issue will be published by CCPE as special issues. A combination weighted clustering algorithm is proposed by Li et al1 in the study to reduce the communication overhead of the distributed cognitive network and maintain the stability of the network structure. Li2 investigates edge-forwarding index of routing in the lexicographic product network, which is used to measure the congestion of an edge. Shang et al3 presents the on-chip evolution system of combinational logic circuits by a new hybrid algorithm known as improved genetic-simulated annealing algorithm. Jianmin et al4 proposes the uniform distribution code, model identification code, standard dimension code, and attribute dimensional code. Aiming at the actual production needs of the Dadingzi mountain hydrojunction, An et al5 studies the optimal control of the flood discharge system. Xu et al6 attempts to analyze the traffic network from the perspective of vulnerability. Ding et al7 provides a basis for adjusting the transport organization plan and evacuating passenger flow in emergency. In order to solve the problems of low utilization of the shipyard unreasonable placed of the blocks and unbalanced use of flat transporters, grid search algorithm for block storage yard scheduling is proposed by Zhou et al.8 Pei et al9 establishes a customer experience evaluation index system from the perspectives of O2O takeout platform, delivery staff, and distribution process. To realize the benign interaction between electric vehicle charging and switching power stations and power grid, and the demand-side management of electric vehicle charging and switching power stations, the Monte Carlo random sampling method is used by Gan and Zheng10 to collect the battery-charged state of batteries in one day. Yang and Liu11 draws on the theory of biological immune system, combines the ontology with the growth risk of start-ups, and uses SWRL and Jess engine to build an ontology knowledge base of start-ups' growth risk to identify various risks in the growth process of start-up. Tran et al12 provides a comprehensive research on cellular network traffic prediction using exponential smoothing methods. Guo and Chen13 proposes an interval support vector domain description method through using the dynamic decreasing inertia weight particle swarm optimization. Xu et al14 present the development of a cloud-based difference detection method using big GPR data for roadbed damage detection. Lu et al15 focuses on textual cyberbullying detection because text is the most common form of social media. The guest editors would like to thank Prof. Geoffrey C. Fox who is the editor in chief of CCPE. His help and trust is the most important thing for the success of this SI. The guest editors would like to thank the reviewers for their high-quality reviews, which provided insightful and constructive feedback to the authors of the articles.