检索结果-内蒙古大学图书馆

38th Conference on Neural Information Processing Systems, NeurIPS 2024

作者： Hu, Zhangyi Yang, Bin Ye, Mang National Engineering Research Center for Multimedia Software School of Computer Science Wuhan University Wuhan China

Visible-Infrared Person Re-identification (VI-ReID) is a challenging cross-modal retrieval task due to significant modality differences, primarily resulting from the absence of color information in the infrared modality. The development of large foundation models like Large Language Models (LLMs) and Vision Language Models (VLMs) motivates us to explore a feasible solution to empower VI-ReID with off-the-shelf large foundation models. To this end, we propose a novel Text-enhanced VI-ReID framework driven by Large Foundation Models (TVI-LFM). The core idea is to enrich the representation of the infrared modality with textual descriptions automatically generated by VLMs. Specifically, we incorporate a pre-trained VLM to extract textual features from texts generated by VLM and augmented by LLM, and incrementally fine-tune the text encoder to minimize the domain gap between generated texts and original visual modalities. Meanwhile, to enhance the infrared modality with extracted textual representations, we leverage modality alignment capabilities of VLMs and VLM-generated feature-level filters. This enables the text model to learn complementary features from the infrared modality, ensuring the semantic structural consistency between the fusion modality and the visible modality. Furthermore, we introduce modality joint learning to align features across all modalities, ensuring that textual features maintain stable semantic representation of overall pedestrian appearance during complementary information learning. Additionally, a modality ensemble retrieval strategy is proposed to leverage complementary strengths of each query modality to improve retrieval effectiveness and robustness. Extensive experiments on three expanded VI-ReID datasets demonstrate that our method significantly improves the retrieval performance, paving the way for the utilization of large foundation models in downstream multi-modal retrieval tasks. © 2024 Neural information processing systems foundat

关键词：

来源：评论

学校读者我要写书评

暂无评论

A novel image encryption algorithm based on new one-dimensional chaos and DNA coding

引用

Multimedia Tools and Applications 2024年第36期83卷 84275-84297页

作者： Feng, Sijia Zhao, Maochang Liu, Zhaobin Li, Yuanyu Tianjin public emergency early warning information release center Tianjin Meteorology Information Center Tianjin300074 China School of Information Science & amp Technology Dalian Maritime University Dalian116026 China School of Computer Science and Software Engineering University of Science and Technology Liaoning Anshan114051 China

This paper introduces a new one-dimensional chaotic system and a new image encryption algorithm. Firstly, the new chaotic system is analyzed. The bifurcation diagram and Lyapunov exponent show that the system has strong chaotic characteristics and is suitable for the field of image encryption. The chaotic sequence generated by the system is used in image encryption, scrambled according to its sequence value, and several sequences required for DNA coding operation are generated. The encrypted image is obtained by encoding, decoding, and diffusion operation according to the sequence. The initial values and parameters of system are generated by Hash algorithm based on plain image, so it has strong sensitivity and can effectively resist attacks such as selective plaintext attack. Experimental results show that the algorithm has high security and robustness, and can resist common attacks. © The Author(s), under exclusive licence to Springer science+Business Media, LLC, part of Springer Nature 2024.

关键词： Chaotic systems

来源：评论

学校读者我要写书评

暂无评论

PKEST: Public-Key Encryption With Similarity Test for Medical Consortia Cloud Computing

引用

IEEE Transactions on Cloud Computing 2025年第2期13卷 680-693页

作者： Chen, Junsong Zeng, Shengke Han, Song Yin, Jin Chen, Peng Xihua University School of Computer and Software Engineering Chengdu610039 China Hangzhou City University School of Computer and Computing Science Hangzhou310015 China Sichuan University West China Biomedical Big Data Center West China Hospital Chengdu610041 China

Cloud computing eliminates the limitations of local hardware architecture while also enabling rapid data sharing between healthcare institutions. Encryption of electronic medical records (EMRs) before uploading to cloud servers is necessary for privacy. However, encryption brings challenges for computation. Public Key Encryption with Equality Test (PKEET) allows cloud servers to test the underlying message equality without decryption. Therefore, it can be used to classify the encrypted EMRs corresponding to different medical symptoms. However, traditional PKEETs have limitations in testing the similarity between the ciphertexts. Undoubtedly, it can not handle EMR classification with similar medical symptoms efficiently. In this work, we propose a lightweight public key encryption with similarity test (PKEST) for the EMR classification shared in medical consortia. Our scheme can resist offline message recovery attacks, which may be launched by the insider manager, and the traditional paring computation is not necessary. Our experiment simulation shows that the similarity error between ciphertext and plaintext is tiny when the parameters are set properly. Compared to previous works, our scheme not only achieves the classification of similar encrypted EMRs but is also more efficient than traditional PKEETs since our construction does not need paring computation anymore. © 2013 IEEE.

关键词： Ciphertext

来源：评论

学校读者我要写书评

暂无评论

Harmonizing space–time dynamics for precision in human action recognition

引用

Multimedia Tools and Applications 2025年第5期84卷 2805-2832页

作者： Majid, Abdul Wang, Yulin Ullah, Anwar Naiz, Fahim Zarar, Muhammad School of Computer Science Wuhan University Hubei Wuhan430072 China National Engineering Research Center for E-Leaning Central China Normal University Hubei Wuhan430072 China

Human action recognition (HAR) plays a vital role in various fields, including surveillance, healthcare, and human–computer interaction. Recognizing multi-actor actions in crowded scenarios poses a significant challenge due to the complex dynamics and interactions among individuals. Existing methods often fail to effectively capture both spatial and temporal dependencies in video frames, leading to inaccuracies in recognizing diverse action classes under real-world conditions. In this paper, we introduce a novel framework called Video Capsule Net-Multi Actor (VCN-MA), designed to enhance the precision of HAR by harmonizing space–time dynamics. The VCN-MA architecture incorporates a deep neural network combining convolutional and recurrent layers to capture spatial and temporal features. Additionally, we leverage pre-trained embeddings and attention mechanisms to focus on critical features during action sequences. The proposed model utilizes capsule networks with dynamic routing to represent complex hierarchical spatial–temporal relationships, further refining action recognition accuracy. We evaluate our approach using the MAMA dataset, which includes 35 diverse action classes under varied real-world conditions such as different viewpoints, occlusions, and lighting variations. The VCN-MA model achieves an overall accuracy of 85% on the test set, outperforming existing state-of-the-art methods. Visualizations, including confusion matrices and ROC curves, along with metrics like the Matthews Correlation Coefficient (MCC), provide a comprehensive analysis of the model's performance across different action classes. These results demonstrate the potential applicability of VCN-MA in practical scenarios, including video surveillance, human–computer interaction, and healthcare applications. © The Author(s), under exclusive licence to Springer science+Business Media, LLC, part of Springer Nature 2024.

关键词： Deep neural networks

来源：评论

学校读者我要写书评

暂无评论

Intelligent Predetermination of Generator Tripping Scheme: Knowledge Fusion-based Deep Reinforcement Learning Framework

引用

CSEE Journal of Power and Energy Systems 2024年第1期10卷 66-75页

作者： Lingkang Zeng Wei Yao Ze Hu Hang Shuai Zhouping Li Jinyu Wen Shijie Cheng State Key Laboratory of Advanced Electromagnetic Engineering and Technology School of Electrical and Electronics EngineeringHuazhong University of Science and TechnologyWuhan 430074China Dispatching and Control Center Central China Branch of State Grid Corporation of ChinaWuhan 430077China. Department of Electrical Engineering and Computer Science University of TennesseeKnoxvilleTN 37996USA

Generator tripping scheme(GTS)is the most commonly used scheme to prevent power systems from losing safety and ***,GTS is composed of offline predetermination and real-time scenario ***,it is extremely time-consuming and labor-intensive for manual predetermination for a large-scale modern power *** improve efficiency of predetermination,this paper proposes a framework of knowledge fusion-based deep reinforcement learning(KF-DRL)for intelligent predetermination of ***,the Markov Decision Process(MDP)for GTS problem is formulated based on transient instability ***,linear action space is developed to reduce dimensionality of action space for multiple controllable ***,KF-DRL leverages domain knowledge about GTS to mask invalid actions during the decision-making *** can enhance the efficiency and learning ***,the graph convolutional network(GCN)is introduced to the policy network for enhanced learning *** simulation results obtained on New England power system demonstrate superiority of the proposed KF-DRL framework for GTS over the purely data-driven DRL method.

关键词： Deep reinforcement learning generator tripping scheme graph convolutional network invalid action masking knowledgefusion

来源：评论

学校读者我要写书评

暂无评论

Belief Combination of Classifiers for Incomplete Data

引用

IEEE/CAA Journal of Automatica Sinica 2022年第4期9卷 652-667页

作者： Zuowei Zhang Songtao Ye Yiru Zhang Weiping Ding Hao Wang IEEE the Research Center for Optical Fiber Sensing Zhejiang Laboratory the School of Automation Northwestern Polytechnical University the School of Cyber Engineering Xidian University the Department of Computer Science CY Cergy Paris Université the School of Information Science and Technology Nantong University the School with of Cyber Engineering Xidian University the Department of Computer Science Norwegian University of Science and Technology

Data with missing values,or incomplete information,brings some challenges to the development of classification,as the incompleteness may significantly affect the performance of *** this paper,we handle missing values in both training and test sets with uncertainty and imprecision reasoning by proposing a new belief combination of classifier(BCC)method based on the evidence *** proposed BCC method aims to improve the classification performance of incomplete data by characterizing the uncertainty and imprecision brought by *** BCC,different attributes are regarded as independent sources,and the collection of each attribute is considered as a ***,multiple classifiers are trained with each subset independently and allow each observed attribute to provide a sub-classification result for the query ***,these sub-classification results with different weights(discounting factors)are used to provide supplementary information to jointly determine the final classes of query *** weights consist of two aspects:global and *** global weight calculated by an optimization function is employed to represent the reliability of each classifier,and the local weight obtained by mining attribute distribution characteristics is used to quantify the importance of observed attributes to the pattern *** comparative experiments including seven methods on twelve datasets are executed,demonstrating the out-performance of BCC over all baseline methods in terms of accuracy,precision,recall,F1 measure,with pertinent computational costs.

关键词： Classifier fusion classification evidence theory incomplete data missing values

来源：评论

学校读者我要写书评

暂无评论

LKAW: A Robust Watermarking Method Based on Large Kernel Convolution and Adaptive Weight Assignment

引用

computers, Materials & Continua 2023年第4期75卷 1-17页

作者： Xiaorui Zhang Rui Jiang Wei Sun Aiguo Song Xindong Wei Ruohan Meng Engineering Research Center of Digital Forensics Ministry of EducationJiangsu Engineering Center of Network MonitoringSchool of Computer and SoftwareNanjing University of Information Science&TechnologyNanjing210044China Wuxi Research Institute Nanjing University of Information Science&TechnologyWuxi214100China Jiangsu Collaborative Innovation Center of Atmospheric Environment and Equipment Technology(CICAEET) Nanjing University of Information Science&TechnologyNanjing210044China School of Automation Nanjing University of Information Science&TechnologyNanjing210044China School of Instrument Science and Engineering Southeast UniversityNanjing211189China School of Teacher Education Nanjing University of Information Science&TechnologyNanjing210044China School of Computer Science Engineering Nanyang Technological UniversitySingapore

Robust watermarking requires finding invariant features under multiple attacks to ensure correct *** learning has extremely powerful in extracting features,and watermarking algorithms based on deep learning have attracted widespread *** existing methods use 3×3 small kernel convolution to extract image features and embed the ***,the effective perception fields for small kernel convolution are extremely confined,so the pixels that each watermarking can affect are restricted,thus limiting the performance of the *** address these problems,we propose a watermarking network based on large kernel convolution and adaptive weight assignment for loss *** uses large-kernel depth-wise convolution to extract features for learning large-scale image information and subsequently projects the watermarking into a highdimensional space by 1×1 convolution to achieve adaptability in the channel ***,the modification of the embedded watermarking on the cover image is extended to more *** the magnitude and convergence rates of each loss function are different,an adaptive loss weight assignment strategy is proposed to make theweights participate in the network training together and adjust theweight ***,a high-frequency wavelet loss is proposed,by which the watermarking is restricted to only the low-frequency wavelet sub-bands,thereby enhancing the robustness of watermarking against image *** experimental results show that the peak signal-to-noise ratio(PSNR)of the encoded image reaches 40.12,the structural similarity(SSIM)reaches 0.9721,and the watermarking has good robustness against various types of noise.

关键词： Robust watermarking large kernel convolution adaptive loss weights high-frequency wavelet loss deep learning

来源：评论

学校读者我要写书评

暂无评论

C2FNet: Coarse-to-Fine Keypoint Localization Network for Monocular 6D Object Pose Estimation

C2FNet: Coarse-to-Fine Keypoint Localization Network for Mon...

引用

2023 China Automation Congress, CAC 2023

作者： Sun, Jiahao Ma, Xin Li, Yibin Center for Robotics School of Control Science and Engineering Shandong University China Engineering Research Center of Intelligent Unmanned System Ministry of Education Jinan China

ISBN: (纸本)9798350303759

Estimating the 6D object pose from a singular RGB image is a fundamental task in the field of computer vision. Recent studies have demonstrated that keypoint-based methods exhibit remarkable efficacy. Such methodologies initially identify keypoints and subsequently deduce the object's pose by addressing a Perspective-n-Point (PnP) problem. Nonetheless., the prevalent approach of directly regressing the 2D coordinates of keypoints renders these methods sensitive to occlusions. Instead., we inaugurate a coarse-to-fine keypoint localization network (C2FNet)., where we utilize the predicted keypoint offsets from the first stage as input to deformable convolutional networks (DCN) in the second stage., to further capture geometric features and enhance spatial relationships between keypoints., thus outputting accurate and robust keypoint coordinates. Experiments indicate that the proposed approach surpasses existing advanced methods in terms of performance on the LINEMOD and Occlusion LINEMOD datasets. © 2023 IEEE.

关键词： computer vision

来源：评论

学校读者我要写书评

暂无评论

A Composite Decomposition Method for Large-Scale Global Optimization

IEEE Transactions on Artificial Intelligence

引用

IEEE Transactions on Artificial Intelligence 2024年第12期5卷 6304-6319页

作者： Tian, Maojiang Chen, Minyang Du, Wei Tang, Yang Jin, Yaochu Yen, Gary G. Southern University of Science and Technology Department of Computer Science and Engineering Shenzhen518055 China East China University of Science and Technology Key Laboratory of Smart Manufacturing in Energy Chemical Process Ministry of Education Shanghai200237 China East China University of Science and Technology Engineering Research Center of Process System Engineering Ministry of Education Shanghai200237 China Westlake University School of Engineering Hangzhou310030 China Oklahoma State University School of Electrical and Computer Engineering StillwaterOK74048 United States

Cooperative coevolution (CC) algorithms, based on the divide-and-conquer strategy, have emerged as the predominant approach to solving large-scale global optimization (LSGO) problems. The efficiency and accuracy of the grouping stage significantly impact the performance of the optimization process. While the general separability grouping (GSG) method has overcome the limitation of previous differential grouping (DG) methods by enabling the decomposition of nonadditively separable functions, it suffers from high computational complexity. To address this challenge, this article proposes a composite separability grouping (CSG) method, seamlessly integrating DG and GSG into a problem decomposition framework to utilize the strengths of both approaches. CSG introduces a step-by-step decomposition framework that accurately decomposes various problem types using fewer computational resources. By sequentially identifying additively, multiplicatively and generally separable variables, CSG progressively groups nonseparable variables by recursively considering the interactions between each nonseparable variable and the formed nonseparable groups. Furthermore, to enhance the efficiency and accuracy of CSG, we introduce two innovative methods: a multiplicatively separable variable detection method and a nonseparable variable grouping method. These two methods are designed to effectively detect multiplicatively separable variables and efficiently group nonseparable variables, respectively. Extensive experimental results demonstrate that CSG achieves more accurate variable grouping with lower computational complexity compared to GSG and state-of-the-art DG-series designs. © 2024 IEEE.

关键词： Computational efficiency

来源：评论

学校读者我要写书评

暂无评论

The propagation of Quasi-static region during granular impact

引用

Particuology 2023年第12期83卷 1-7页

作者： Moyu Yang Ran Li Yingnan Xiu Qiang Zeng Xiaoyan Ye Hui Yang School of Optical-Electrical and Computer Engineering University of Shanghai for Science and TechnologyShanghai200093China Center for Particle-Laden Turbulence Key Laboratory of Mechanics on Disaster and Environment in Western China attached to the Ministry of Education of ChinaDepartment of Mechanics and Engineering ScienceSchool of Civil Engineering and MechanicsLanzhou UniversityLanzhou730000China

In this paper,through experiments and DEM simulations,it is found that there is a ring-shaped region named Quasi-static Region between the particles expanding outward and the particles collapsing inward during the impact ***-static Region is always generated from the impact point at the same time,and then spreads out at a uniform *** the propagation of Quasi-static Region,the velocity of the particles in the particle extend region varies linearly in space and *** the change rate is only related to the properties of the particle *** simulation results show that the particle flow in Quasi-static Region is elastic-inertial flow,while the particle flow expanding outward and collapsing inward is inertial flow.

关键词： Impact Quasi-staticregion Propagation Shearflow

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：