检索结果-内蒙古大学图书馆

Underwater Image Quality Evaluation: A Comprehensive Review

IET Image Processing 2025年第1期19卷

作者： Shen, Mengjiao Yang, Miao Zhong, Jinyang Liu, Hantao Pan, Can School of Electronic Engineering Jiangsu Ocean University Lianyungang China School of Computer Science and Informatics Cardiff University Cardiff United Kingdom

Underwater image quality evaluation (UIQE) is crucial in improving image processing techniques and optimizing the design of the imaging system to obtain object information more accurately. However, existing UIQE methods are designed based on limited images or consider only a few natural scene statistics (NSS) metrics, lacking consideration for generalization across various underwater imaging applications. In this paper, an in-depth review of the existing UIQE methods based on evaluation operations is provided, emphasizing the bias present when evaluating UIQE methods using individual metrics. To address this, a novel metric called quadrilateral datum evaluation (QDE) is designed for UIQE methods. It comprehensively considers robustness across different datasets, as well as correlation and ranking consistency with mean opinion scores (MOS). This is the first solution to measure an UIQE method from an all-encompassing visual perspective. By using QDE, UIQE methods characterized by greater feature strength and small imbalance demonstrate good consistency and robustness across multiple aspects, providing a basis for the design of UIQE methods. © 2025 The Author(s). IET Image Processing published by John Wiley & Sons Ltd on behalf of The Institution of Engineering and Technology.

关键词： Image correlation

来源：评论

学校读者我要写书评

暂无评论

Multi-level Matching Network for Multimodal Entity Linking 25

Multi-level Matching Network for Multimodal Entity Linking

引用

Proceedings of the 31st ACM SIGKDD Conference on Knowledge Discovery and Data Mining V.1

作者： Zhiwei Hu Victor Gutiérrez-Basulto Ru Li Jeff Z. Pan School of Computer and Information Technology Shanxi University Taiyuan China School of Computer Science and Informatics Cardiff University Cardiff United Kingdom ILCC School of Informatics University of Edinburgh Edinburgh United Kingdom

ISBN: (纸本)9798400712456

Multimodal entity linking (MEL) aims to link ambiguous mentions within multimodal contexts to corresponding entities in a multimodal knowledge base. Most existing approaches to MEL are based on representation learning or vision-and-language pre-training mechanisms for exploring the complementary effect among multiple modalities. However, these methods suffer from two limitations. On the one hand, they overlook the possibility of considering negative samples from the same modality. On the other hand, they lack mechanisms to capture bidirectional cross-modal interaction. To address these issues, we propose a Multi-level Matching network for Multimodal Entity Linking(M3EL). Specifically, M3EL is composed of three different modules: (i) a Multimodal Feature Extraction module, which extracts modality-specific representations with a multimodal encoder and introduces an intra-modal contrastive learning sub-module to obtain better discriminative embeddings based on uni-modal differences; (ii) an Intra-modal Matching Network module, which contains two levels of matching granularity: Coarse-grained Global-to-Global and Fine-grained Global-to-Local, to achieve local and global level intra-modal interaction; (iii) a Cross-modal Matching Network module, which applies bidirectional strategies, Textual-to-Visual and Visual-to-Textual matching, to implement bidirectional cross-modal interaction. Extensive experiments conducted on WikiMEL, RichpediaMEL, and WikiDiverse datasets demonstrate the outstanding performance of M3EL when compared to the state-of-the-art baselines.

关键词： data mining

来源：评论

学校读者我要写书评

暂无评论

LATUP-Net: A lightweight 3D attention U-Net with parallel convolutions for brain tumor segmentation

引用

computers in Biology and Medicine 2025年 184卷 109353-109353页

作者： Alwadee, Ebtihal J. Sun, Xianfang Qin, Yipeng Langbein, Frank C. School of Computer Science and Informatics Cardiff University CardiffCF24 4AG United Kingdom Department of Computer Science College of Engineering and Computer Science Jazan University Jazan Saudi Arabia

Early-stage 3D brain tumor segmentation from magnetic resonance imaging (MRI) scans is crucial for prompt and effective treatment. However, this process faces the challenge of precise delineation due to the tumors’ complex heterogeneity. Moreover, energy sustainability targets and resource limitations, especially in developing countries, require efficient and accessible medical imaging solutions. The proposed architecture, a Lightweight 3D ATtention U-Net with Parallel convolutions, LATUP-Net, addresses these issues. It is specifically designed to reduce computational requirements significantly while maintaining high segmentation performance. By incorporating parallel convolutions, it enhances feature representation by capturing multi-scale information. It further integrates an attention mechanism to refine segmentation through selective feature recalibration. LATUP-Net achieves promising segmentation performance: the average Dice scores for the whole tumor, tumor core, and enhancing tumor on the BraTS 2020 dataset are 88.41%, 83.82%, and 73.67%, and on the BraTS 2021 dataset, they are 90.29%, 89.54%, and 83.92%, respectively. Hausdorff distance metrics further indicate its improved ability to delineate tumor boundaries. With its significantly reduced computational demand using only 3.07 M parameters, about 59 times fewer than other state-of-the-art models, and running on a single NVIDIA GeForce RTX3060 12 GB GPU, LATUP-Net requires just 15.79 GFLOPs. This makes it a promising solution for real-world clinical applications, particularly in settings with limited resources. Investigations into the model's interpretability, utilizing gradient-weighted class activation mapping and confusion matrices, reveal that while attention mechanisms enhance the segmentation of small regions, their impact is nuanced. Achieving the most accurate tumor delineation requires carefully balancing local and global features. The code is available at https://***/ca/code-bca. © 2024 T

关键词： Magnetic resonance imaging

来源：评论

学校读者我要写书评

暂无评论

FedCSpc: A Cross-Silo Federated Learning System with Error-Bounded Lossy Parameter Compression

引用

IEEE Transactions on Parallel and Distributed Systems 2025年第7期36卷 1372-1386页

作者： Zhang, Zhaorui Di, Sheng Zhao, Kai Jin, Sian Tao, Dingwen Ji, Zhuoran Liu, Benben Alharthi, Khalid Ayed Cao, Jiannong Cappello, Franck The Hong Kong Polytechnic University Department of Computing Hong Kong Florida State University Department of Computer Science United States Temple University Department of Computer and Information Sciences United States Indiana University Bloomington Luddy School of Informatics Computing and Engineering United States Shandong University School of Cyber Science and Technology China The University of Hong Kong LSCM R&D Center Hong Kong University of Bisha Department of Computer Science College Of Computing And Information Technology Saudi Arabia Argonne National Laboratory United States

Cross-Silo federated learning is widely used for scaling deep neural network (DNN) training over data silos from different locations worldwide while guaranteeing data privacy. Communication has been identified as the main bottleneck when training large-scale models due to large-volume model parameters and gradient transmission across public networks with limited bandwidth. Most previous works focus on gradient compression, while limited work tries to compress parameters that can not be ignored and extremely affect communication performance during the training. To bridge this gap, we propose FedCSpc: an efficient cross-silo federated learning system with an XAI-driven adaptive parameter compression strategy for large-scale model training. Our work substantially differs from existing gradient compression techniques due to the distinct data features of gradient and parameter. The key contributions of this paper are fourfold. (1) Our designed FedCSpc proposes to compress the parameter during the training using the state-of-the-art error-bounded lossy compressor – SZ3. (2) We develop an adaptive compression error bound adjustment algorithm to guarantee the model accuracy effectively. (3) We exploit an efficient approach to utilize the idle CPU resources of clients to compress the parameters. (4) We perform a comprehensive evaluation with a wide range of models and benchmarks on a GPU cluster with 65 GPUs. Results show that FedCSpc can achieve the same model accuracy as FedAvg while reducing the data volume of parameters and gradients in communication by up to 7.39× and 288×, respectively. With 32 clients on a 4Gb size model, FedCSpc significantly outperforms FedAvg in wall-clock time in the emulated WAN environment (at the bandwidth of 1 Gbps or lower without loss of generality). © 1990-2012 IEEE.

关键词： Bandwidth compression

来源：评论

学校读者我要写书评

暂无评论

Diversity, Equity, and Inclusion in Computing science: Culture is the Key, Curriculum Contributes 2024

Diversity, Equity, and Inclusion in Computing Science: Cultu...

引用

29th Working Group Reports on Innovation and Technology in computer science Education, ITiCSE WGR 2024

作者： Toti, Giulia Lindner, Peggy Gao, Alice Baghban Karimi, Ouldooz Engineer, Rutwa Hur, Jinyoung Mcneill, Fiona Reckinger, Shanon Robinson, Rebecca Sollazzo, Anna Wicentowski, Richard Department of Computer Science University of British Columbia VancouverBC Canada Department of Information Science Technology University of Houston HoustonTX United States Department of Computer Science University of Toronto TorontoON Canada School of Computing Science Simon Fraser University SurreyBC Canada Department of Mathematical and Computational Sciences University of Toronto Mississauga MississaugaON Canada Department of Computer Science University of Illinois Urbana-Champaign Urbana-ChampaignIL United States School of Informatics University of Edinburgh Edinburgh United Kingdom Department of Computer Science University of Illinois Chicago ChicagoIL United States Department of Human Centred Computing Monash University ClaytonVIC Australia School of Computing Science University of Glasgow United Kingdom Computer Science Department Swarthmore College SwarthmorePA United States

ISBN: (纸本)9798400712081

Undergraduate computer science programs worldwide struggle to attract and retain underrepresented students for many reasons. Culture, stereotype threats, uneven gender and racial representations, lack of role models, and uncertain career prospects for minority groups are among the many reasons behind this situation. Many computer science programs are trying to change course through strategies to foster equity, diversity, and inclusion (EDI), aimed at improving outreach, recruitment, admissions, and retention of underrepresented students. EDI approaches may also include modifications to the undergraduate computer science curriculum. However, if not properly planned, these modifications risk amplifying existing stereotypes rather than producing positive change [38]. In this study, through an extensive literature review, a rigorous curriculum analysis of 49 computer science programs across the globe, and qualitative and quantitative analysis of surveys and interviews bringing in the voices of 613 students and 30 educators participating from around the world, we explore equity, diversity, and inclusion in the computer science curriculum. We highlight the role of inclusive content and course design, discuss program flexibility, and the impact of inclusive courses and program design in attracting and retaining historically marginalized students. Finally, we provide concrete steps to make computing science undergraduate curricula more appealing to a diverse audience. © 2024 Copyright held by the owner/author(s).

关键词： Students

来源：评论

学校读者我要写书评

暂无评论

MatrikalinDiabetes: User-Centered Design of a mHealth App for Gestational Diabetes Mellitus Management and Education Among Bangladeshi Women 26th

MatrikalinDiabetes: User-Centered Design of a mHealth App ...

引用

26th International Conference on Human-computer Interaction, HCII 2024

作者： Ratul, Mohammad Arshad Hossain Yanoor Bristy, Tunisha Sayeed, Noorjahan Islam, Ashraful Chaudhry, Beenish Moalla Center for Computational and Data Sciences Independent University Bangladesh Dhaka Bangladesh Department of Computer Science and Engineering Independent University Bangladesh Dhaka Bangladesh School of Computing and Informatics University of Louisiana at Lafayette Louisiana United States

ISBN: (纸本)9783031768088

This study employs the User-Centered Design (UCD) methodology to develop a mobile health (mHealth) application (app) specifically tailored for Bangladeshi women with Gestational Diabetes Mellitus (GDM). GDM affects approximately 10% of pregnant women globally and around 35% within Bangladesh, as reported by the World Health Organization (WHO). Despite its high prevalence, there is a significant lack of awareness and education about GDM in Bangladesh, compounded by language barriers that make existing digital solutions less accessible. The MatrikalinDiabetes mHealth app aims to overcome these barriers by providing comprehensive management and educational resources for GDM in the Bangla language, addressing both pre-pregnancy and post-pregnancy needs. To ensure the app meets the actual needs of potential users, personas were created based on a literature review, and a survey was conducted. This process informed the development of both a low-fidelity, paper-based prototype and a high-fidelity digital prototype. Key features of the MatrikalinDiabetes app include tracking of food and water intake, physical activity monitoring, reminders, GDM education, chat forums for community support, blood glucose level (BGL) monitoring, an SOS button for emergency contacts, and quizzes for user engagement. Preliminary feedback indicated that 53.8% of participants favored the integrated features and expressed willingness to use and recommend the app. The survey of the high-fidelity prototype yielded positive responses, with 57.9% of participants strongly agreeing and 42.1% agreeing that the features were beneficial. The user-friendly design of the high-fidelity prototype ensures intuitive interaction, aiding in health management, enhancing GDM understanding, and promoting a healthier lifestyle among pregnant women in Bangladesh. © The Author(s), under exclusive license to Springer Nature Switzerland AG 2025.

关键词： Obstetrics

来源：评论

学校读者我要写书评

暂无评论

Objective Metrics for Human-Subjects Evaluation in Explainable Reinforcement Learning

arXiv

引用

arXiv 2025年

作者： Gyevnar, Balint Towers, Mark School of Informatics University of Edinburgh United Kingdom School of Electronics and Computer Science University of Southampton United Kingdom

Explanation is a fundamentally human process. Understanding the goal and audience of the explanation is vital, yet existing work on explainable reinforcement learning (XRL) routinely does not consult humans in their evaluations. Even when they do, they routinely resort to subjective metrics, such as confidence or understanding, that can only inform researchers of users’ opinions, not their practical effectiveness for a given problem. This paper calls on researchers to use objective human metrics for explanation evaluations based on observable and actionable behaviour to build more reproducible, comparable, and epistemically grounded research. To this end, we curate, describe, and compare several objective evaluation methodologies for applying explanations to debugging agent behaviour and supporting human-agent teaming, illustrating our proposed methods using a novel grid-based environment. We discuss how subjective and objective metrics complement each other to provide holistic validation and how future work needs to utilise standardised benchmarks for testing to enable greater comparisons between research. © 2025, CC BY.

关键词： Contrastive Learning

来源：评论

学校读者我要写书评

暂无评论

PRIVACY-PRESERVING GENERATIVE MODELS: A COMPREHENSIVE SURVEY

arXiv

引用

arXiv 2025年

作者： Padariya, Debalina Wagner, Isabel Taherkhani, Aboozar Boiten, Eerke School of Computer Science and Informatics De Montfort University Leicester United Kingdom Department of Mathematics and Computer Science University of Basel Basel Switzerland

Despite the generative model’s groundbreaking success, the need to study its implications for privacy and utility becomes more urgent. Although many studies have demonstrated the privacy threats brought by GANs, no existing survey has systematically categorized the privacy and utility perspectives of GANs and VAEs. In this article, we comprehensively study privacy-preserving generative models, articulating the novel taxonomies for both privacy and utility metrics by analyzing 100 research publications. Finally, we discuss the current challenges and future research directions that help new researchers gain insight into the underlying concepts. Copyright © 2025, The Authors. All rights reserved.

关键词： Differential privacy

来源：评论

学校读者我要写书评

暂无评论

DRDM: A Disentangled Representations Diffusion Model for Synthesizing Realistic Person Images

DRDM: A Disentangled Representations Diffusion Model for Syn...

引用

2025 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2025

作者： Huang, Enbo Zhang, Yuan Huang, Faliang Zhang, Guangyu Liu, Yang Guangxi Key Lab of Human-machine Interaction and Intelligent Decision Nanning Normal University Nanning China School of Computer and Information Engineering Nanning Normal University Nanning China College of Mathematics and Informatics South China Agricultural University Guangzhou China School of Computer Science and Engineering Sun Yat-sen University Guangzhou China

ISBN: (纸本)9798350368741

Person image synthesis with controllable body poses and appearances is an essential task owing to the practical needs in the context of virtual try-on, image editing and video production. However, existing methods face significant challenges with details missing, limbs distortion and the garment style deviation. To address these issues, we propose a Disentangled Representations Diffusion Model (DRDM) to generate photo-realistic images from source portraits in specific desired poses and appearances. First, a pose encoder is responsible for encoding pose features into a high-dimensional space to guide the generation of person images. Second, a body-part subspace decoupling block (BSDB) disentangles features from the different body parts of a source figure and feeds them to the various layers of the noise prediction block, thereby supplying the network with rich disentangled features for generating a realistic target image. Moreover, during inference, we develop a parsing map-based disentangled classifier-free guided sampling method, which amplifies the conditional signals of texture and pose. Extensive experimental results on the Deepfashion dataset demonstrate the effectiveness of our approach in achieving pose transfer and appearance control. The associated project can be found at https://***/lovemusiceb/DRDM. © 2025 IEEE.

关键词： Diffusion Model Disentangled Representation Human Parsing Person image synthesis

来源：评论

学校读者我要写书评

暂无评论

ZKCross: An Efficient and Reliable Cross-Chain Authentication Scheme Based on Lightweight Attribute-Based Zero-Knowledge Proof 24th

ZKCross: An Efficient and Reliable Cross-Chain Authenticati...

引用

24th International Conference on Algorithms and Architectures for Parallel Processing, ICA3PP 2024

作者： Xu, Yuwei Cai, Hailang Chen, Jialuo Xiang, Qiao Xu, Jingdong Cheng, Guang School of Cyber Science and Engineering Southeast University Nanjing China Purple Mountain Laboratories for Network and Communication Security Nanjing China Engineering Research Center of Blockchain Application Supervision and Management Nanjing China School of Informatics Xiamen University Xiamen China College of Computer Science Nankai University Tianjin China

ISBN: (纸本)9789819615506

The consortium blockchain finds wide application in finance and logistics. However, these independent networks create information silos. Cross-chain authentication aims to bridge these gaps. It verifies qualifications across different blockchains, yet it can risk leaking sensitive data. The main methods include Attribute-Based Signatures (ABS) and Attribute-Based Zero-Knowledge Proof (ABZKP). ABZKP better handles key management, general applicability, and attribute verification than ABS. However, the current ABZKP suffers from large proof files and slow verification. Additionally, many schemes can’t handle sets of multiple attributes and may introduce centralization. Overall, the current schemes based on zero-knowledge proof still struggle to strike a balance between security and efficiency. We propose a reliable cross-chain authentication scheme. Our lightweight zero-knowledge proof algorithm enhances verification speed and reduces file size. We also introduce parallel verification to address security concerns. Security analysis and experiments show that ZKCross when properly configured, achieves near-perfect accuracy and outperforms others in size and speed. Thus, ZKCross stands as an efficient and dependable cross-chain authentication scheme. © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2025.

关键词： Authentication Protocol

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：