ABSTRACT Recent 3D point cloud classification has predominantly focused on local spatial attention, neglecting distant contextual relationships due to the inherent sparsity of LiDAR‐generated data over longer distances. Existing 3D object detection methods prioritize local features, hindering the extraction of semantic information. Despite attempts with transformers, methods often reduce computations through local spatial attention, neglecting content class and scarcely establishing connections among distant global points. Our proposed point class‐adaptive transformer (PCaT) addresses these limitations by establishing long‐range feature dependencies while significantly reducing computations. PCaT includes three key modules: the class‐adaptive transformer (CaT), which utilizes local self‐attention and global self‐attention based on class similarity to facilitate an efficient trade‐off between capturing extended‐global dependencies and managing computational challenges; nested binary clustering (NbC), which dynamically partitions queries into multiple clusters based on content features in each Transformer block; and the AfA, which aggregates high‐dimensional features using max‐pooling alongside a residual MLP component and low‐dimensional features using average pooling and a CaT block. Additionally, PCaT incorporates point cloud segmentation via local–global feature aggregation (PcSeg) to facilitate effective point cloud segmentation. Extensive experimentation on the ModelNet40, ScanObjectNN, and S3DIS datasets demonstrates the superior performance and reasonable stability of PCaT compared with existing methods. PCaT achieves 94.2% overall accuracy (OA) and mIoU scores of 89.2% and 86.2% for the ScanObjectNN and S3DIS datasets, respectively.

3D object detection by autonomous vehicles is integral to intelligent transportation. Existing systems often compromise essential foreground point features and local spatial interactions through random down-sampling, focusing primarily on local feature extraction. However, this neglects interactions among distant yet significant points, limiting semantic information and detection performance due to inherent point cloud data sparsity. Addressing this, our proposed Geometric Feature Aggregation and Self-Attention in a Multi-Head Transformer (GFA-SMT) architecture leverages Graph Convolutional Networks and multi-channel transformers to enhance weak semantic information of distant sparse objects. GFA-SMT comprises three modules: Distance Suppression for Local Receptive Fields (DsLRF), Geometric Feature Aggregator with Multi-head Self Attention (GFaSA), and Predicted Key-point Weighting and Refinement (PKwR). DsLRF preserves foreground features, GFaSA encodes similar features and aggregates edge features, while PKwR focuses on key-points for enhancing geometric knowledge of distant and sparse objects. Extensive experiments on KITTI, DIARV2X-I and NuScenes datasets show significant enhancements in widely used techniques, resulting in notable increases in average precision (AP) for 3D object detection: 4.08%, 5.56%, and 4.62%, respectively, on the KITTI test dataset. GFA-SMT enhances point cloud detection accuracy, particularly at medium and long distances, with minimal impact on run-time performance and model parameters.

In the modern digital age, technological advancements have led to unprecedented connectivity, notably with the rise of connected networks that integrate vehicles into the Internet of Vehicles (IoV), thereby heightening the risk of cyber threats. Network Intrusion Detection Systems (NIDS) are crucial for protecting these systems by identifying and mitigating unauthorized access. Traditional machine learning techniques, though effective, often struggle with sophisticated and evolving threats. This study introduces an Optimized Ensemble approach for enhanced intrusion detection in the IoV. Our approach, the Optimized Random Forest (Opt-Forest), combines Decision Forest approaches with Genetic Algorithms (GAs) to improve detection accuracy. Feature selection methods, including Best-First Search, Particle Swarm Optimization, Evolutionary Search, and Genetic Search, are employed to boost the model’s adaptability and resilience against modern threats. We evaluated our approach against established machine learning models like K-Nearest Neighbor (KNN), J48-Decision Tree (J48), and Multilayer Perceptron (MLP). The results demonstrate the superior performance of our approach across various metrics, highlighting its potential to significantly enhance network intrusion detection in the IoV environment.

Current 3D object detection methods face limitations in both single-modal and multimodal approaches. Single-modal detectors struggle with inadequate depth perception or difficulty distinguishing semantically similar objects, while multimodal systems, which combine LiDAR and camera data, encounter challenges like integration complexity and inefficient feature fusion. This study presents ROI-based ViT Fusion (RVF3D), a multimodal 3D object detector combining sparse 3D and dense 2D data, outperforming current methods. The RVF3D enhances the understanding of rich LiDAR and camera representations through improved query generation, feature sampling, and multimodality cross-ViT fusion. We propose an ROI-based Query Sampling (RQS) multimodal 3D object identification pipeline, eliminating laborious non-maximum suppression (NMS) postprocessing and complex prior box configurations. The RQS module uses learnable filters to aggregate image and point representations, retaining foreground characteristics for object localization and recognition while removing background noise. Our RVF3D leverages a hierarchical vision transformer-based approach, ViT-Fusion, which includes CameraViT and LidarViT components for embedding representations of input data in each modality. These representations are fused to enable hierarchical learning and deeper feature extraction. Comprehensive tests on the KITTI and nuScenes datasets demonstrate that RVF3D achieves 89.88% mAP in BEV and 75.2% mAP in 3D, respectively.

In computer vision, accurately detecting objects in three-dimensional (3D) scenes is indispensable for intelligent transportation, robotics, vision applications, and augmented reality. While self-attention mechanisms have demonstrated remarkable success in enhancing feature representations and capturing long-range dependencies, their application to 3D object detection remains challenging due to inherent limitations to capturing intricate contextual dependencies among points to refining 3D proposals. This paper proposes a high-quality Chunk-wise Reweighting Transformer for 3D Object Detection (CRT3D) architecture explicitly tailored for 3D object detection tasks. CRT3D comprises three modules: self-attention-based encoder-decoder for effective modelling of spatial context; channel-based Reweighting captures temporal dependencies; and Chunk-wise Reweighting improves computational efficiency to efficiently handle sparse point cloud data, advancing real-time 3D object detection in autonomous vehicles. Our CRT3D model can effectively balance the trade-off between capturing local and global context, leading to more informative and contextually rich representations. Through extensive experiments on the benchmark dataset KITTI [1], we demonstrate the efficacy of CRT3D in achieving better performance by 3.49% and 4.25% in 3D AP with PointPillar and PointRCNN backbone, respectively. CRT3D enhances 3D object detection framework accuracy, particularly at medium range and occlusion cases, while minimally impacting real-time performance and model parameters.

Current 3D detection methods, whether single-modal or multimodal, face notable limitations. Single-modal detectors, using either camera or LiDAR, struggle with spatial accuracy and object differentiation due to insufficient depth information or difficulty distinguishing semantically similar objects. Existing multimodal fusion techniques, while improving performance, often suffer from high computational costs, false positives, and complex architectures, especially when utilizing anchor-based pipelines. To address these challenges, we propose an efficient pointwise fusion method that directly extracts point features from enhanced RGB images and fuses them with corresponding point cloud features, preserving essential spatial and semantic information. This fused data is then processed through a three-dimensional neural network, significantly improving inference speed and detection performance. Our framework is designed for multi-class 3D object detection, leveraging the complementary strengths of LiDAR and camera data without the need for multiple backbones or complex synchronization steps. Extensive experiments on the KITTI benchmark demonstrate that the proposed method outperforms state-of-the-art LiDAR-camera fusion techniques, achieving 92.5% AP for 3D detection and 95.41% AP for BEV detection, making it particularly suitable for autonomous driving systems. These results highlight the effectiveness of the proposed fusion strategy in balancing accuracy, computational efficiency, and robustness in complex 3D environments.

The growing vulnerabilities in computer networks highlight the critical need for robust intrusion detection techniques, particularly host-based systems. This paper evaluates five host-based intrusion detection systems (HIDS) to identify the most effective solution. Among these, the distributed intrusion detection system (DIDS) stands out due to its dual functionality, addressing threats at both local hosts and network levels for enhanced security. Computer applications, comprising interconnected hardware and tools, enable efficient send data and applications and data sharing but are face to various threats. Network threats, such as machine tampering, electrical sparks, and environmental factors, can compromise network infrastructure. Likewise, software threats, including malware and prohibitted access, often exploit hidden vulnerabilities, posing severe risks to data integrity and system performance. Therefore, this study emphasizes proactive measures, including secure setups, regular monitoring, good services, redundant power lines, and controlled environmental conditions. High-level threat detection and vulnerability management are essential for software threats. By implementing these approaches, target-based systems can reduce both hardware and software threats, ensuring network security and scalability. The results underscore the importance of preemptive tasks to identify and address vulnerabilities and improve the resilience of host-based intrusion detection systems in safeguarding modern computer networks against evolving threats.

Accurate 3-D object detection is vital in autonomous driving. Traditional LiDAR models struggle with sparse point clouds. We propose a novel approach integrating LiDAR and camera data to maximize sensor strengths while overcoming individual limitations for enhanced 3-D object detection. Our research introduces the channelwise and spatially guided multimodal feature fusion network (CSMNET) for 3-D object detection. First, our method enhances LiDAR data by projecting it onto a 2-D plane, enabling the extraction of class-specific features from a probability map. Second, we design class-based farthest point sampling (C-FPS), which boosts the selection of foreground points by utilizing point weights based on geometric or probability features while ensuring diversity among the selected points. Third, we developed a parallel attention (PAT)-based multimodal fusion mechanism achieving higher resolution compared to raw LiDAR points. This fusion mechanism integrates two attention mechanisms: channel attention for LiDAR data and spatial attention for camera data. These mechanisms enhance the utilization of semantic features in a region of interest (ROI) to obtain more representative point features, leading to a more effective fusion of information from both LiDAR and camera sources. Specifically, CSMNET achieves an average precision (AP) in bird’s eye view (BEV) detection of 90.16% (easy), 85.18% (moderate), and 80.51% (hard), with a mean AP (mAP) of 85.12%. In 3-D detection, CSMNET attains 82.05% (easy), 72.64% (moderate), and 67.10% (hard) with an mAP of 73.75%. For 2-D detection, the scores are 95.47% (easy), 93.25% (moderate), and 86.68% (hard), yielding an mAP of 91.72% for the KITTI dataset.

Abstract Software Architectural Process (SAP) is a core and excessively knowledge intensive phase of software development life cycle, as it consumes and produces knowledge artifacts, simultaneously. SAP is about making design decisions, and the changes in these verdicts may pose adverse effects on software projects. The performance and properties of software components are fundamentally influenced by the design decisions. The implementation of immature and abrupt design decisions seriously threatens the development process of SAP. Moreover, software architectural knowledge management (AKM) approaches offer systematic ways to support SAP through versatile architectural solutions and design decisions. However, the majority of software organizations have limited access to data and still depend upon manually created and maintained AKM process. In this paper, we have utilized the one of the most prominent online community for software development (i.e., Stack Overflow) as a source of SAP knowledge to support AKM. In order to support AKM, we have proposed a supervised machine learning‐based approach to classify the architectural knowledge into predefined categories, that is, analysis, synthesis, evaluation, and implementation. We have employed different combinations of feature selection technique to achieve the optimal classification results of the used classifiers (Support Vector Machine [SVM], K‐Nearest Neighbor, Random Forest, and Naive Bayes [NB]). Among these classifiers, SVM with Uni‐gram feature set provides best classification results and attains 85.80% accuracy. For evaluating the proposed approach's effectiveness, we have also computed the suitability of the classifiers, that is, the cost of computation along with its accuracy, and NB with Uni‐gram feature set proved to be the most suitable.