检索结果-内蒙古大学图书馆

32nd Wireless and Optical Communications Conference, WOCC 2023

作者： Lee, Ho-Cheng Lin, Fuchun Joseph Chen, Jyh-Cheng Chen, Chien Wang, Patrick Department of Computer Science College of Computer Science Nycu Hsinchu Taiwan College of Computer Science Nycu Communication Service/Software Laboratory Hsinchu Taiwan

ISBN: (纸本)9798350337150

This research employs NYCU-developed open source 5G core network free5GC and Intel open source edge computing platform OpenNESS to build a 5G private network. An online multi-person chorus application is then deployed on the edge platform to (1) achieve High Reliability and Low Latency Communication (URLLC) requirements, and (2) improve the backhaul bandwidth occupancy rate from the edge to the core network. In addition, this research implements the traffic influence function proposed in the 3GPP 5G standards, which can dynamically change traffic rules of the 5G core during execution, directing specific traffic to the edge applications in order to improve the performance of private networks. Finally, to verify the effectiveness of this schema, this research uses the example application deployed to compare the performance of the system equipped with edge computing with that without edge platform. Our analysis is done with both a physical RAN and the UERANSIM simulator. © 2023 IEEE.

关键词： 5G mobile communication systems

来源：评论

学校读者我要写书评

暂无评论

CDF-Net: A zero-shot 3D classification network with CLIP Decoder for Feature Fusion 24

CDF-Net: A zero-shot 3D classification network with CLIP Dec...

引用

2024 International Conference on Intelligent Perception and Pattern Recognition, IPPR 2024

作者： Yan, Hao Bai, Jing Liu, Lu School of Computer Science and Engineering North Minzu University China The Key Laboratory of images Graphics Intelligent Processing of State Ethnic Affairs Commission China

ISBN: (纸本)9798400707315

Zero-shot 3D model classification requires neural networks to correctly classify models with unseen categories, which has important practical implications for fields such as autonomous driving and robotics. PointCLIP demonstrates that transferring knowledge from the 2D image domain to the 3D model domain is beneficial for zero-shot 3D model classification, even without incorporating new 3D-specific knowledge. However, its effectiveness is limited by the lack of 3D knowledge. As a result, subsequent improvements have focused on refining 3D-to-2D projection algorithms rather than exploiting 3D information. Our approach aims to exploit both image domain knowledge and the full potential of 3D models. To this end, we propose CDF-Net, a learnable model architecture. CDF-Net consists of two components: an initial feature encoding module using the Contrastive Language-Image Pre-training network to extract 2D knowledge from multi-view representations of 3D models, and a learnable module to extract 3D knowledge and integrate it with 2D knowledge. Evaluations on two benchmarks consisting of four datasets show that CDF-Net achieves competitive classification accuracies, demonstrating the effectiveness of our method. © 2024 Copyright held by the owner/author(s).

关键词： Zero-shot learning

来源：评论

学校读者我要写书评

暂无评论

P2LNet: HD Map Validation Using Graph Neural Networks 1

P2LNet: HD Map Validation Using Graph Neural Networks

引用

1st International Conference on Robotics, Engineering, science, and Technology, RESTCON 2024

作者： Reji, Jeevan Omanwar, Vaibhav Birla Institute of Technology & Sciences Pilani Department of Computer Science Hyderabad Campus India Nvidia Graphics Pvt Ltd Pune India

ISBN: (纸本)9798350348842

HD Maps are a highly important part of the autonomous driving stack to perform the activities of localization and route planning on the road. Therefore, HD Map Validation is crucial to ensure the HD Maps represent the road information as accurately as possible. While heuristics have been developed to validate HD Maps with reasonable accuracy, they are unable to solve the challenging path to traffic light validation problem of HD Map junctions. Based on the success of Graph Neural Network approaches on HD Map problems, such as Vector net, we propose the P2LNet. This P2LNet architecture consists of a fully connected subgraph followed by a Graph Encoder-Decoder Architecture that finally predicts the correct associations that would exist between paths and traffic lights in the junction. We trained and evaluated P2LNet on our in-house HD Map junction dataset, with P2LNet showing 94% accuracy on predicting the correct labels for the associations from a test set. While the results could be further improved by using edge features, P2LNet provides a significant accuracy in validating incorrect light associations. It also shows how GNN based approaches can be used to solve other significant HD Map and validation issues. © 2024 IEEE.

关键词： Autonomous vehicles

来源：评论

学校读者我要写书评

暂无评论

Endogenous Security of FQn Networks: Adaptive System-Level Fault Self-Diagnosis

引用

IEEE Transactions on Reliability 2024年第3期73卷 1659-1668页

作者： Lin, Yuhang Lin, Limei Huang, Yanze Xu, Li Hsieh, Sun-Yuan Fujian Normal University College of Computer and Cyber Security Fujian Provincial Key Laboratory of Network Security and Cryptology Fuzhou350117 China Fujian University of Technology School of Computer Science and Mathematics Fuzhou350118 China National Cheng Kung University Department of Computer Science and Information Engineering Tainan701 Taiwan

Endogenous security has the ability to discover, eliminate, and solve internal security problems and hidden dangers within the network, and is a superior technology to ensure future network security. The t/k-diagnosis strategy, as a strong and adaptive self-diagnosis strategy, is an important part for ensuring endogenous security. Moreover, the folded hypercube (FQn) as a data transmission network (e.g., optical network) topology offers new potential for the construction of large-scale, high data throughput, and low-latency systems, such as computing network, human-cyber-physical systems, and smart grid. However, there are few studies on endogenous security based on FQn networks. Therefore, this article designs an adaptive system-level fault self-diagnosis strategy, namely Fast t/k-Diagnosis Under Maeng-Malek Model (Ftk-DIAG-MM*) to diagnosis the faulty vertices in FQn network under the Maeng-Malek model (MM∗ mod). Then, we provide a proof of the algorithm correctness theoretically by the fault tolerance of FQn network. It is derived by theoretical derivation that the t/k-diagnosability inherent to the FQn network is (n+1)(k+1)-k(k+3)/2. The simulation experiments demonstrate that the designed Ftk-DIAG-MM∗ strategy can correctly diagnose all vertices within the range allowed by the diagnosability, and still has a great performance when it exceeds the range allowed by the diagnosability. It greatly enhances the fault diagnosis capability of FQn network in the circumstance of misdiagnosing a few vertices, which provides an important theoretical basis for the reliability and endogenous security of FQn networks. © 1963-2012 IEEE.

关键词： Topology

来源：评论

学校读者我要写书评

暂无评论

Text Categorization Using Convolutional and Bidirectional Fast Gated Recurrent Unit 12th

Text Categorization Using Convolutional and Bidirectional Fa...

引用

12th International Conference on Information Systems and Advanced Technologies, ICISAT 2022

作者： Assia, Belherazem Redouane, Tlemsani SIMPA Laboratory Computer Science Department University of Science and Technology Oran Algeria Computer Science Department University of Science and Technology Oran Algeria

ISBN: (纸本)9783031253430

A novel supervised model is presented in this paper with One-Dimensional Convolutional Neural Networks (CNNs) followed by double Bidirectional Gated Recurrent Units (2-BiGRUs) for text categorization. The overall model contains six components, two of which are required. Starting with the input layer which converts the text into an embedding matrix and then passed to the embedding layer via GloVe and Word2Vec embeddings. The convolution block then extracts the bidirectional temporal characteristics from text data and flatting them by the Flatten layer, using Softmax activation and dropouts;the results are then normalized using the Batch Normalization layer and sent to the Dense or output Layer which generates the categorization results by using the Softmax activation. Multiple experiments with loss and precision scores are the primary criterion for evaluating the proposed model on WebKb, Reuters R8 and R52, AG-News (AG), and 20 NewsGroups (20NG) datasets. Our model achieves better precision results on the WebKB, AG, and 20 NG data sets than the best results of other approaches. The proposed model demonstrates that performance scores are superior. © 2023, The Author(s), under exclusive license to Springer Nature Switzerland AG.

关键词： Bidirectional GRU Categorization CNN Deep learning Multiclass categorization Word embedding

来源：评论

学校读者我要写书评

暂无评论

N-Gram Based HASSANIYA Dialect Classification 6

N-Gram Based HASSANIYA Dialect Classification

引用

6th International Conference on Intelligent Computing in Data sciences, ICDS 2024

作者： El Arby Chrif, Mohamed El Moustapha Saleck, Fatimetou Mohamed Nanne, Mohamedade Farouk Beqqali, Omar El Mahmoud, El Benany Mohamed University Sidi Mohamed Ben Abdallah LISAC Laboratory Department of Computer Science Fez Morocco Nouakchott University CSIDS Department of Mathematics and Computer Science Nouakchott Mauritania

ISBN: (纸本)9798350351200

This paper focuses on the HASSANIYA dialect, which consists of the same Arabic letters and is widely used locally as a Mauritania dialect. Despite other local dialects, it has also been studied in terms of sentiment analysis. Yet, it still needs further studies to understand how word roots could be extracted and provide clear directions for future research in the relevant axis of Natural Language Processing (NLP), as text classification by machine learning techniques is largely influenced by the affixes of words, which could change the meaning of words. In this work, we present a novel approach that aims to study the effect of the N-gram method on a HASSANIYA dataset containing around two thousand comments collected from the Facebook platform and evaluate how it influenced the text classification results after preprocessing them. Additionally, three machine learning techniques, such as Support Vector Machine (SVM), Random Forests (RF), and Logistic Regression (LR), were applied for evaluation. The experiments give satisfactory results, confirming that the unigram method performs better than the bigram method for all algorithms. © 2024 IEEE.

关键词： Adversarial machine learning

来源：评论

学校读者我要写书评

暂无评论

Fine-grained point cloud classification based on adaptive KNN algorithm 24

Fine-grained point cloud classification based on adaptive KN...

引用

2024 International Conference on Intelligent Perception and Pattern Recognition, IPPR 2024

作者： Liu, Lu Bai, Jing Yan, Hao School of Computer Science and Engineering North Minzu University China The Key Laboratory of images Graphics Intelligent Processing of State Ethnic Affairs Commission China

ISBN: (纸本)9798400707315

In point cloud deep learning, local feature aggregation is crucial to improve the performance of the model. The traditional KNN (K-Nearest Neighbors) method performs local feature aggregation by fixing the K value, but often performs poorly in the face of different local structural variations. In order to overcome these limitations, this study proposes an improved KNN method that adapts to the diversity of local regions of the point cloud by adaptively adjusting the neighborhood size in order to maintain feature diversity and accuracy in regions of varying importance. Specifically, in this paper, we dynamically adjust different neighborhood ranges according to the attention weights of individual points, thus ensuring that local features are effectively captured in both important as well as non-important regions. In addition, we conduct experiments on the FG3D and ModelNet40, point cloud datasets. The overall accuracy of AK-Net reaches 96.31%, 79.85%, and 82.38% in the experimental results on the three sub-datasets Airplane, Car, and Chair of FG3D, and 94.04% on ModelNet40. Therefore, the classification performance of AK-Net is superior. © 2024 Copyright held by the owner/author(s).

关键词： Adaptive algorithms

来源：评论

学校读者我要写书评

暂无评论

DDR: Exploiting Deep Degradation Response as Flexible Image Descriptor 38

DDR: Exploiting Deep Degradation Response as Flexible Image ...

引用

38th Conference on Neural Information Processing Systems, NeurIPS 2024

作者： Wu, Juncheng Ni, Zhangkai Wang, Hanli Yang, Wenhan Zhou, Yuyin Wang, Shiqi School of Computer Science and Technology Tongji University China Pengcheng Laboratory China Department of Computer Science and Engineering University of California Santa Cruz United States Department of Computer Science City University of Hong Kong Hong Kong

Image deep features extracted by pre-trained networks are known to contain rich and informative representations. In this paper, we present Deep Degradation Response (DDR), a method to quantify changes in image deep features under varying degradation conditions. Specifically, our approach facilitates flexible and adaptive degradation, enabling the controlled synthesis of image degradation through text-driven prompts. Extensive evaluations demonstrate the versatility of DDR as an image descriptor, with strong correlations observed with key image attributes such as complexity, colorfulness, sharpness, and overall quality. Moreover, we demonstrate the efficacy of DDR across a spectrum of applications. It excels as a blind image quality assessment metric, outperforming existing methodologies across multiple datasets. Additionally, DDR serves as an effective unsupervised learning objective in image restoration tasks, yielding notable advancements in image deblurring and single-image super-resolution. Our code is available at: https://***/eezkni/DDR. © 2024 Neural information processing systems foundation. All rights reserved.

关键词：

来源：评论

学校读者我要写书评

暂无评论

SECOND ORDER DIFFERENTIAL INCLUSION WITH UNBOUNDED NONCONVEX MOVING SETS

引用

UPB Scientific Bulletin, Series A: Applied Mathematics and Physics 2023年第4期85卷 67-78页

作者： Mecemma, Imene Lounis, Sabrina Yarou, Mustapha Fateh Laboratory of LMPA Department of Mathematics Faculty of exact sciences and computer science University of Jijel Jijel18000 Algeria

We prove, in this work, an existence result for a class of second order nonconvex sweeping processes depending on both time and state, subject to unbounded external forces known as perturbations, in infinite dimensional Hilbert spaces. The approach is based on Moreau’s catching-up algorithm. In order to deal with a large class of unbounded nonconvex sets, we assume a truncated condition. An application for a class of quasi-variational inequality is given. © 2023, Politechnica University of Bucharest. All rights reserved.

关键词： Hilbert spaces

来源：评论

学校读者我要写书评

暂无评论

Text-Guided Portrait Image Matting

IEEE Transactions on Artificial Intelligence

引用

IEEE Transactions on Artificial Intelligence 2024年第8期5卷 4149-4162页

作者： Xu, Yong Yao, Xin Liu, Baoling Quan, Yuhui Ji, Hui South China University of Technology School of Computer Science and Engineering Guangzhou510000 China Communication and Computer Network Laboratory of Guangdong Guangzhou510641 China Peng Cheng Laboratory Shenzhen518000 China National University of Singapore Department of Mathematics 119076 Singapore

Image matting is a technique used to separate the foreground of an image from the background, which estimates an alpha matte that indicates pixel-wise degree of transparency. To precisely extract target objects and address the ambiguity of solutions in image matting, many existing approaches employ a trimap or background image provided by the user as additional input to guide the matting process. This article introduces a novel matting paradigm termed text-guided image matting, utilizing a textual description of the foreground object as a guiding element. In contrast to trimap or background-based methods, text-guided matting offers a user-friendly interface, providing semantic clues for the objects of interest. Moreover, it facilitates batch processing across multiple frames featuring the same objects of interest. The proposed text-guided matting approach is implemented through a deep neural network (NN) comprising three-stage cross-modal feature fusion and two-step alpha matte prediction. Experimental results on portrait matting demonstrate the competitive performance of our text-guided approach compared to existing trimap-based and background-based methods. © 2020 IEEE.

关键词： Data mining

来源：评论

学校读者我要写书评

暂无评论

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案：

请选择收藏分类：

通借通还

建议与咨询 留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案： 新增检索档案 确定 取消

请选择收藏分类： 新增自定义分类 确定 取消

通借通还

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

请选择保存的检索档案：

请选择收藏分类：