检索结果-内蒙古大学图书馆

TSGFormer: temporal-aware network and spatial encoding GCN for three-dimensional human pose estimation

Multimedia Systems 2025年第3期31卷 1-19页

作者： Xiao, Xinwang Zhao, Huihuang Li, Yuhang Tang, Peng Deng, Yue College of Computer Science and Technology Hengyang Normal University Hengyang421008 China Hunan Provincial Key Laboratory of Intelligent Information Processing and Application Hengyang Normal University Hengyang421008 China

Transformer-based approaches have significantly driven recent progress in three-dimensional human pose estimation. However, existing transformer-based approaches are still deficient in capturing localized features, and they lack task-specific a priori information by obtaining queries, keys, and values through simple linear mappings. Existing methods lack effective human constraints for model training. We introduce the Spatial Encoding Graph Convolutional Network Transformer (SEGCNFormer), designed to enhance model capacity in capturing local features. In addition, we propose a Temporal-Aware Network, which generates queries, keys, and values possessing a priori knowledge of human motion, enabling the model to better understand the structural information of human poses. Finally, we leverage the knowledge of human anatomy and motion to design the Human Structural Science Loss, which performs a rationality assessment of human actions and imposes physical constraints on the generated poses. Our method outperforms existing methods on the Human3.6M dataset in both 27 and 81 sampling frames, and our predicted poses are closer to the actual poses with less error. For the existing three issues, we proposed effective methods and conducted targeted experiments, which confirmed the effectiveness of our strategies. © The Author(s), under exclusive licence to Springer-Verlag GmbH Germany, part of Springer Nature 2025.

关键词： Signal encoding

来源：评论

学校读者我要写书评

暂无评论

Broad-Learning-System-Based Model-Free Adaptive Predictive Control for Nonlinear MASs Under DoS Attacks

引用

IEEE/CAA Journal of Automatica Sinica 2025年第2期12卷 381-393页

作者： Hongxing Xiong Guangdeng Chen Hongru Ren Hongyi Li the School of Automation Guangdong-Hong Kong Joint Laboratory for Intelligent Decision and Cooperative Control Guangdong University of Technology Guangdong Provincial Key Laboratory for Intelligent Decision and Cooperative Control Guangdong University of Technology the College of Electronic and Information Engineering Southwest University IEEE

In this paper, the containment control problem in nonlinear multi-agent systems(NMASs) under denial-of-service(DoS) attacks is addressed. Firstly, a prediction model is obtained using the broad learning technique to train historical data generated by the system offline without DoS attacks. Secondly, the dynamic linearization method is used to obtain the equivalent linearization model of NMASs. Then, a novel model-free adaptive predictive control(MFAPC) framework based on historical and online data generated by the system is proposed, which combines the trained prediction model with the model-free adaptive control method. The development of the MFAPC method motivates a much simpler robust predictive control solution that is convenient to use in the case of DoS attacks. Meanwhile, the MFAPC algorithm provides a unified predictive framework for solving consensus tracking and containment control problems. The boundedness of the containment error can be proven by using the contraction mapping principle and the mathematical induction method. Finally, the proposed MFAPC is assessed through comparative experiments.

关键词： Broad learning technique denial-of-service (DoS) model-free adaptive predictive control (MFAPC) nonlinear multi-agent systems (NMASs)

来源：评论

学校读者我要写书评

暂无评论

RECONSTRUCTION-BASED A POSTERIORI ERROR ESTIMATES FOR THE L1 METHOD FOR TIME FRACTIONAL PARABOLIC PROBLEMS

引用

Journal of Computational Mathematics 2025年第2期43卷 345-368页

作者： Jiliang Cao Aiguo Xiao Wansheng Wang Department of Mathematics Shanghai Normal UniversityShanghai 200234China Hunan Key Laboratory for Computation and Simulation in Science and Engineering&Key Laboratory of Intelligent Computing and Information Processing of Ministry of Education Xiangtan UniversityXiangtan 411105China

In this paper,we study a posteriori error estimates of the L1 scheme for time discretizations of time fractional parabolic differential equations,whose solutions have generally the initial *** derive optimal order a posteriori error estimates,the quadratic reconstruction for the L1 method and the necessary fractional integral reconstruction for the first-step integration are *** using these continuous,piecewise time reconstructions,the upper and lower error bounds depending only on the discretization parameters and the data of the problems are *** numerical experiments for the one-dimensional linear fractional parabolic equations with smooth or nonsmooth exact solution are used to verify and complement our theoretical results,with the convergence ofαorder for the nonsmooth case on a uniform *** recover the optimal convergence order 2-αon a nonuniform mesh,we further develop a time adaptive algorithm by means of barrier function recently *** numerical implementations are performed on nonsmooth case again and verify that the true error and a posteriori error can achieve the optimal convergence order in adaptive mesh.

关键词： Time fractional parabolic differential equations A posteriori error estimates Ll method Fractional integral reconstruction Quadratic reconstruction

来源：评论

学校读者我要写书评

暂无评论

Turbulent drag reduction by sector-shaped counter-flow dielectric barrier discharge plasma actuator

引用

Chinese Physics B 2025年第2期34卷 362-370页

作者： Borui Zheng Shaojie Qi Minghao Yu Jianbo Zhang Linwu Wang Dongliang Bian School of Automation and Information Engineering Xi’an University of TechnologyXi’an 710048China School of Mechanical and Precision Instrument Engineering Xi’an University of TechnologyXi’an 710048China Science and Technology on Plasma Dynamics Laboratory Air Force Engineering UniversityXi’an 710038China Shaanxi Key Laboratory of Complex System Control and Intelligent Information Processing Xi’an 710048China Shaanxi University Key Laboratory of Photonic Power Devices and Discharge Regulation Xi’an 710048China

The primary objective in aircraft transportation is to minimize turbulent drag, thereby conserving energy and reducing emissions. We propose a sector-shaped counter-flow dielectric barrier discharge plasma actuator, which leverages jet synthesis for drag reduction. A drag control experiment was conducted in a low-speed wind tunnel with a controlled flow velocity of 9.6 m/s(Re = 1.445 × 10^(4)). This study investigated the effects of varying pulse frequencies and actuation voltages on the turbulent boundary layer. Using a hot-wire measurement system, we analyzed the pulsating and time-averaged velocity distributions within the boundary layer to evaluate the streamwise turbulent drag reduction. The results show that the local TDR decreases as the pulse frequency increases, reaching a maximum reduction of approximately 20.97% at a pulse frequency of 50 Hz. In addition, as the actuation voltage increases, the friction coefficient decreases, increasing the drag reduction rate. The maximum drag reduction of approximately 33.34% is achieved at an actuation voltage of 10 kV.

关键词： plasma flow control turbulent boundary layer turbulent drag reduction

来源：评论

学校读者我要写书评

暂无评论

Wireless Charging Scheduling for Long-term Utility Optimization

引用

ACM Transactions on Sensor Networks 2025年第1期21卷 1-31页

作者： Xu, Jia Chen, Wenbin Dai, Haipeng Xu, Lijie Xiao, Fu Liu, Linfeng Jiangsu Key Laboratory of Big Data Security and Intelligent Processing Nanjing University of Posts and Telecommunications Nanjing China The State Key Laboratory for Novel Software Technology Nanjing University Nanjing China

Wireless power transmission has been widely used to replenish energy for wireless sensor networks, where the energy consumption rate of sensor nodes is usually time varying and indefinite. However, few works have investigated the problem of long-term charging scheduling with random variable. This article designs an optimization model for the long-term scheduling of chargers to maximize the time-averaged charging utility while ensuring its time-averaged constraints of budget and response rate. The Lyapunov optimization technique is adopted to transform the stochastic optimization problem into a deterministic optimization problem, which remains NP-hard. Thus, an approximation algorithm following greedy approach is proposed to solve the deterministic optimization problem. We further provide the theoretical analysis of feasibility and performance guarantee of the proposed scheduling algorithm. The simulation results show that our algorithm outperforms three comparison algorithms by 6.53%, 20.04%, and 19.97% in terms of time-averaged charging utility, as well as by 11.25%, 4.42%, and 3.73% in terms of time-averaged response rate on average. © 2025 Copyright held by the owner/author(s). Publication rights licensed to ACM.

关键词： Inductive power transmission

来源：评论

学校读者我要写书评

暂无评论

HyFit: Hybrid Fine-Tuning With Diverse Sampling for Abstractive Summarization

引用

IEEE Transactions on Big Data 2025年第3期11卷 1054-1065页

作者： Zhao, Shu Cheng, Yuanfang Zhang, Yanping Chen, Jie Duan, Zhen Sun, Yang Wang, Xinyuan Anhui University Ministry of Education Key Laboratory of Intelligent Computing and Signal Processing Hefei230601 China Anhui University School of Computer Science and Technology Hefei230601 China Anhui University Information Materials and Intelligent Sensing Laboratory of Anhui Province Hefei230601 China

Abstractive summarization has made significant progress in recent years, which aims to generate a concise and coherent summary that contains the most important facts from the source document. Current fine-tuning approaches based on pre-training models typically rely on autoregressive and maximum likelihood estimation, which may result in inconsistent historical distributions generated during the training and inference stages, i.e., exposure bias problem. To alleviate this problem, we propose a hybrid fine-tuning model(HyFit), which combines contrastive learning and reinforcement learning in a diverse sampling space. Firstly, we introduce reparameterization and probability-based sampling methods to generate a set of summary candidates called candidates bank, which improves the diversity and quality of the decoding sampling space and incorporates the potential for uncertainty. Secondly, hybrid fine-tuning with sampled candidates bank, upweighting confident summaries and downweighting unconfident ones. Experiments demonstrate that HyFit significantly outperforms the state-of-the-art models on SAMSum and DialogSum. HyFit also shows good performance on low-resource summarization, on DialogSum dataset, using only approximate 8% of the examples exceed the performance of the base model trained on all examples. © 2015 IEEE.

关键词： Reinforcement learning

来源：评论

学校读者我要写书评

暂无评论

Self-distilled Dual-Network with Pixel Screening Loss for Blind Image Deblurring 27th

Self-distilled Dual-Network with Pixel Screening Loss for ...

引用

27th International Conference on Pattern Recognition, ICPR 2024

作者： Li, Tianyi Tian, Ming Gao, Changxin Sang, Nong National Key Laboratory of Multispectral Information Intelligent Processing Technology School of Artificial Intelligence and Automation Huazhong University of Science and Technology Wuhan China

ISBN: (纸本)9783031781247

In the field of image processing, blind image deblurring aims to restore sharp details in images blurred by an unknown convolution kernel. Recent advancements have shown that deep networks can act as effective image generative priors (DIP) for restoring clear images without requiring external datasets. However, the inherent non-uniqueness of solutions in blind image deblurring often leads DIP-based methods to converge on local optima, resulting in over- or under-deblurred images. To overcome this limitation, we propose a novel deblurring framework featuring dual image generators. These generators mutually constrain each other during training, guiding the model towards the optimal solution. Building on our network structure, we employ a self-ensemble and self-distillation strategy to guide network training, further enhancing performance. Additionally, we introduce a novel loss function based on a pixel screening method, which focuses on the important pixels. This loss enables the network to model the blur kernel more accurately and facilitates the restoration of image details. Our experiments demonstrate that our deblurring approach outperforms most existing methods both qualitatively and quantitatively. © The Author(s), under exclusive license to Springer Nature Switzerland AG 2025.

关键词： Restoration

来源：评论

学校读者我要写书评

暂无评论

Deep learning for parameter identification of nonlinear dynamical system driven by fractional Brownian motion

引用

Nonlinear Dynamics 2025年 1-20页

作者： Hou, Wentao Ma, Shaojuan School of Mathematics and Information Science North Minzu University Ningxia Yinchuan750021 China Department of Mathematics and Information Technology Yuncheng University Shanxi Yuncheng044000 China Ningxia Key Laboratory of Intelligent Information and Big Data Processing Ningxia Yinchuan750021 China

Accurate parameters identification in nonlinear dynamical systems is crucial for understanding complex phenomena. Fractional Brownian motion is widely used in modeling nonlinear systems due to its long memory, but it also complicates parameter identification. Traditional methods typically rely on linear models or assume Gaussian white noise, which makes them less effective for handling such complex systems. Although data-driven methods have been proposed, most of them utilize simple network architectures and focus primarily on identifying the Hurst exponent or parameters of linear systems. This paper presents a deep learning-based parameter identifier capable of jointly identifying all parameters of a nonlinear dynamical system driven by fractional Brownian motion. Firstly, a new parameter identifier (Transformer-Bidirectional Long Short-Term Memory Neural Network) is constructed by combining the advantages of Transformer in dealing with long-term dependence and Bidirectional Long Short-Term Memory in extracting local features. We then incorporate fractional Brownian motion as the random effect in the nonlinear dynamical system and provide three examples to simulate the system. Finally, the proposed identifier is used to identify parameters in the three systems individually, and its performance is compared to that of the Convolutional Neural Network-Bidirectional Long Short-Term Memory-Attention Neural Network and the Parameter Estimation Neural Network to demonstrate its effectiveness. Taking Example 3 and using the mean absolute error to evaluate identification accuracy. The mean absolute error values for the Transformer-Bidirectional Long Short-Term Memory Neural Network are 0.0171, 0.0699, 0.0328 and 0.0851, which are significantly better than the Convolutional Neural Network-Bidirectional Long Short-Term Memory-Attention Neural Network, with values of 0.0277, 0.1054, 0.0494 and 0.1037, and the Parameter Estimation Neural Network, with values of 0.0419, 0.1265,

关键词： Brownian movement

来源：评论

学校读者我要写书评

暂无评论

Deep Reinforcement Learning-based Multi-Objective Scheduling for Distributed Heterogeneous Hybrid Flow Shops with Blocking Constraints

引用

Engineering 2025年第3期46卷 278-291页

作者： Xueyan Sun Weiming Shen Jiaxin Fan Birgit Vogel-Heuser Fandi Bi Chunjiang Zhang State Key Laboratory of Intelligent Manufacturing Equipment and Technology Huazhong University of Science and TechnologyWuhan 430074China Institute of Automation and Information Systems Technical University of MunichGarching 85748Germany

This paper investigates a distributed heterogeneous hybrid blocking flow-shop scheduling problem(DHHBFSP)designed to minimize the total tardiness and total energy consumption simultaneously,and proposes an improved proximal policy optimization(IPPO)method to make real-time decisions for the DHHBFSP.A multi-objective Markov decision process is modeled for the DHHBFSP,where the reward function is represented by a vector with dynamic weights instead of the common objectiverelated scalar value.A factory agent(FA)is formulated for each factory to select unscheduled jobs and is trained by the proposed IPPO to improve the decision *** FAs work asynchronously to allocate jobs that arrive randomly at the shop.A two-stage training strategy is introduced in the IPPO,which learns from both single-and dual-policy data for better data *** proposed IPPO is tested on randomly generated instances and compared with variants of the basic proximal policy optimization(PPO),dispatch rules,multi-objective metaheuristics,and multi-agent reinforcement learning *** experimental results suggest that the proposed strategies offer significant improvements to the basic PPO,and the proposed IPPO outperforms the state-of-the-art scheduling methods in both convergence and solution quality.

关键词： Multi-objective Markov decision process Multi-agent deep reinforcement learning Proximal policy optimization Distributed hybrid flow-shop scheduling Blocking constraints

来源：评论

学校读者我要写书评

暂无评论

A Unified Framework for Adversarial Patch Attacks against Visual 3D Object Detection in Autonomous Driving

引用

IEEE Transactions on Circuits and Systems for Video technology 2025年第5期35卷 4949-4962页

作者： Wang, Jian Li, Fan He, Lijun Xi'an Jiaotong University Shaanxi Key Laboratory of Deep Space Exploration Intelligent Information Technology School of Information and Communications Engineering Xi'an710049 China

The rapid development of vision-based 3D perceptions, in conjunction with the inherent vulnerability of deep neural networks to adversarial examples, motivates us to investigate realistic adversarial attacks for the 3D detection models in autonomous driving scenarios. Due to the perspective transformation from 3D space to the image and object occlusion, current 2D image attacks are difficult to generalize to 3D detectors and are limited by physical feasibility. In this work, we propose a unified framework to generate physically printable adversarial patches with different attack goals: 1) instance-level hiding - pasting the learned patches to any target vehicle allows it to evade the detection process;2) scene-level creating - placing the adversarial patch in the scene induces the detector to perceive plenty of fake objects. Both crafted patches are universal, which can take effect across a wide range of objects and scenes. To achieve above attacks, we first introduce the differentiable image-3D rendering algorithm that makes it possible to learn a patch located in 3D space. Then, two novel designs are devised to promote effective learning of patch content: 1) a Sparse Object Sampling Strategy is proposed to ensure that the rendered patches follow the perspective criterion and avoid being occluded during training, and 2) a Patch-Oriented Adversarial Optimization is used to facilitate the learning process focused on the patch areas. Both digital and physical-world experiments are conducted and demonstrate the effectiveness of our approaches, revealing potential threats when confronted with malicious attacks. We also investigate the defense strategy using adversarial augmentation to further improve the model's robustness. © 1991-2012 IEEE.

关键词： Digital elevation model

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：