检索结果-内蒙古大学图书馆

DCT-DiffPose: A Lightweight Diffusion Model With Multi-Hypothesis for 3D Human Pose Estimation

IEEE Access 2025年 13卷 73319-73331页

作者： Zhong, Linzhan Chen, Fangni Zheng, Bolun Feng, Rui Zhang, Lei Wan, Jian Zhejiang University of Science and Technology College of Information and Electronic Engineering Hangzhou310023 China Zhejiang Key Laboratory of Biomedical Intelligent Computing Technology Hangzhou310023 China Hangzhou Dianzi University School of Automation Hangzhou310018 China

3D human pose estimation is a crucial task in computer vision with extensive applications, yet it remains challenging due to depth ambiguity and constraints on computational efficiency. In this paper, we propose DCT-DiffPose, a novel framework that integrates a diffusion model with Confidence and Consistency-based Multi-Hypothesis Aggregation (CCMA). Moreover, it incorporate the Discrete Cosine Transform (DCT) for frequency-domain feature extraction. Specifically, the diffusion model generates diverse and plausible hypotheses, and CCMA aggregates them based on confidence and consistency, effectively addressing depth ambiguity. Additionally, we incorporate DCT into the self-attention mechanism to transform input data into the frequency domain, thereby enhancing feature extraction while significantly reducing computational complexity. To validate DCT-DiffPose, we conducted extensive experiments on the Human3.6M and MPI-INF-3DHP datasets. Our method achieves a 19% lower Mean Per Joint Position Error (MPJPE) and a 55% reduction in FLOPs compared to D3DP. The results demonstrate its excellent trade-off between accuracy and complexity. © 2013 IEEE.

关键词： Discrete cosine transforms

来源：评论

学校读者我要写书评

暂无评论

GCINet: global convolution interaction network with a pre-trained reversible normalization method for long-term time series forecasting

引用

Neural computing and Applications 2025年第5期37卷 3227-3244页

作者： Fan, Jin Yang, Baoshun Sun, Danfeng Liu, Jie Chen, Qikai Wang, Zhipeng Wu, Jia Hangzhou Dianzi University Hangzhou China Key Laboratory of Intelligent Education Technology and Application of Zhejiang Province Zhejiang Normal University Hangzhou China School of Computing Macquarie University Sydney Australia

Long-term time series forecasting is crucial in various domains, such as weather forecasting and disease forecasting. With the rise of Transformer-based models, we have witnessed a significant improvement in long-term prediction accuracy, as these models use self-attention to capture long-term temporal features in the data during feature extraction. However, this approach has several drawbacks: Firstly, self-attention has quadratic complexity, and reducing its complexity often results in reduced prediction accuracy. Secondly, the distribution of time series data frequently changes over time, a phenomenon known as distribution shift, which is not taken into consideration in most Transformer-based models. This leads to decreased prediction accuracy as the distribution deviates from the one used for model training. Therefore, we have developed a new time series forecasting model, GCINet, which introduces a new mode of interaction based on a convolutional network. This captures both long-term and short-term temporal features while reducing complexity during feature extraction. Additionally, GCINet includes an Inverse Trained Normalization method to address distribution shift, bringing a fresh perspective to the design of time series forecasting models. Our experiments on long-term forecasting tasks demonstrate that GCINet outperforms other commonly used models. © The Author(s), under exclusive licence to Springer-Verlag London Ltd., part of Springer Nature 2024.

关键词： Prediction models

来源：评论

学校读者我要写书评

暂无评论

Surface-Continuous Scene Representation for Light Field Depth Estimation via Planarity Prior

引用

IEEE Transactions on Circuits and Systems for Video technology 2025年第5期35卷 5051-5066页

作者： Chen, Rongshan Sheng, Hao Yang, Da Cui, Zhenglong Cong, Ruixuan Beihang University State Key Laboratory of Virtual Reality Technology and Systems School of Computer Science and Engineering Beijing100191 China Beihang University Key Laboratory of Data Science and Intelligent Computing Hangzhou International Innovation Institute Zhejiang Hangzhou311115 China Macao Polytechnic University Faculty of Applied Sciences China

Light field (LF) imaging captures both spatial and angular information of the real world, enabling precise depth estimation. However, images are merely discrete expressions of scenes. Limited by imaging technology, LF camera cannot capture the infinite rays emitted by scenes, leading to the discrete information storage (e.g. pixel). Consequently, previous deep learning methods have encountered challenges in accurately extracting depth information from LF images. In this paper, we investigate a surface-continuous scene representation using planarity prior and design PlaneNet, a Plane-based Network that successfully generates highly detailed depth maps for real scenes. Specifically, inspired by the plane assumption that real-world scenes generally yield piecewise smooth surfaces, we refine it to the pixel level for continuous surface approximation, which can overcome the limitations of discrete representation. Rather than explicitly parameterizing planes as multiple coefficients, we propose a novel plane regular sampling operator (PRSO), enabling the network to fit smooth depth surfaces easily. To explore the role of our theory at the feature level, we also introduce PRSO into the intermediate layers of PlaneNet. Experiments show that our method achieves state-of-the-art performance on both synthetic and real-world LF scenes, ranking 1st (MSE) on the HCI 4D Light Field benchmark. Furthermore, we explore the utilization of our representation in multiple LF depth estimation networks, and experiments demonstrate improved performance when surface-continuous representation is applied. © 1991-2012 IEEE.

关键词： Pixels

来源：评论

学校读者我要写书评

暂无评论

Multi-granularity feature intersection learning for visible-infrared person re-identification

引用

Complex and intelligent Systems 2025年第7期11卷 1-14页

作者： Chan, Sixian Wang, Jie Cui, Jiaao Hu, Jie Li, Zhuorong Mao, Jiafa College of Computer Science and Technology Zhejiang University of Technology Hangzhou 310023 China Key Laboratory of Intelligent Informatics for Safety & Emergency of Zhejiang Province Wenzhou University Wenzhou 325035 China School of Computer and Computing Science Hangzhou City University Hangzhou 310015 China

This paper proposes a multi-granularity feature intersection network (MGFINet) for visible-infrared person re-identification (VI-ReID). VI-ReID aims to retrieve images of the same pedestrian from different spectral cameras. The key challenge is to extract pedestrian descriptions with both inter-class discriminability and intra-class similarity. Previous methods ignore the potential loss of details during representation extraction and the presence of data bias in the metric function, limiting further improvements in retrieval performance. Meanwhile, the discrepancy regarding how to calculate the loss for representation learning and metric learning also affects the model’s training. To address the above issues, MGFINet consists of three components: a hierarchical part pooling method (HPP), a hierarchical part restriction method (HPC), and a feature intersection (FI) loss. HPP adopts a hierarchical framework to extract multi-granularity pedestrian representations, and it performs an inter-layer fusion operation to exploit the high-resolution information from shallow layers and the semantic representability from deep layers. Meanwhile, HPP employs part pooling with different step sizes to capture pedestrian details in each layer. Next, HPC spreads the identity loss across all layers to reduce the distance for gradient backpropagation and further optimize fine-grained features in shallow layers. Besides, FI loss combines representation and metric learning by incorporating hyperparameters of classifiers into metric learning, mitigating data bias and reducing the gap between the two learning processes. Finally, extensive experiments evaluated on two public datasets, SYSU-MM01 and RegDB demonstrate the effectiveness of the proposed method. © The Author(s) 2025.

关键词： Metric learning Multi-granularity Representation learning VI-ReID

来源：评论

学校读者我要写书评

暂无评论

GoPIM: GCN-Oriented Pipeline Optimization for PIM Accelerators 31

GoPIM: GCN-Oriented Pipeline Optimization for PIM Accelerato...

引用

31st IEEE International Symposium on High Performance Computer Architecture, HPCA 2025

作者： Yang, Siling He, Shuibing Wang, Wenjiong Yin, Yanlong Wu, Tong Chen, Weijian Zhang, Xuechen Sun, Xian-He Feng, Dan The State Key Laboratory of Blockchain and Data Security Zhejiang University China Zhejiang Lab China Institute of Blockchain and Data Security China Zhejiang Key Laboratory of Big Data Intelligent Computing China Washington State University Vancouver United States Illinois Institute of Technology United States Huazhong University of Science and Technology China Wuhan National Laboratory for Optoelectronics China

ISBN: (纸本)9798331506476

Graph convolutional networks (GCNs) are popular for a variety of graph learning tasks. ReRAM-based processing-in-memory (PIM) accelerators are promising to expedite GCN training owing to their in-situ computing capability. However, existing accelerators can be severely underutilized even with pipelines, due to the oversight of the skewed execution times of various GCN stages and the ignorance of skewed degrees of graph vertices. In this work, we propose GOPIM, a GCN-oriented pipeline optimization for PIM accelerators to expedite GCN training. First, GOPIM proposes an ML-based scheme that allocates crossbar resources to the most needed stages to streamline the overall pipeline. Second, GOPIM utilizes a selective vertex updating technique that evenly distributes vertices on crossbars by interleaved mapping. These techniques collectively reduce the overall execution time without losing much accuracy. We also provide a practical architecture design for GOPIM. Our experimental results show that, GoPIM achieves up to 191 × speedup and 16.1 × energy saving, compared to the state-of-the-art work. © 2025 IEEE.

关键词： Graph neural networks

来源：评论

学校读者我要写书评

暂无评论

A Dual-Capture and Dual-Output 3D DNA Walker System Integrated with Ligases Enables Ultrasensitive Detection of Single-Nucleotide Polymorphisms

引用

Analytical Chemistry 2025年第17期97卷 9316-9325页

作者： Zhang, Yunshan Tan, Qianglong Yang, Fang Huang, Tuo Yu, Siyu Ye, Jing Zeng, Jianxian Feng, Xianzhong Zhang, Diming Key Laboratory of Soybean Molecular Design Breeding National Key Laboratory of Black Soils Conservation and Utilization Northeast Institute of Geography and Agroecology Chinese Academy of Sciences Changchun130102 China Research Center for Novel Computing Sensing and Intelligent Processing Zhejiang Laboratory Hangzhou311121 China Hangzhou Institute for Advanced Study University of Chinese Academy of Sciences Zhejiang Hangzhou310024 China School of Chemistry and Chemical Engineering Hunan University of Science and Technology Hunan Xiangtan411201 China

DNA walkers, as structurally and functionally programmable signal amplification tools, exhibit great potential for application in the field of biosensing. Traditional DNA walkers often rely on enzymes for operation, posing compatibility challenges, while the handful of existing enzyme-free DNA walkers demonstrate limited performance. To address this, we innovatively developed an efficient enzyme-free 3D DNA walker with dual capture and dual output capabilities. Coupled with ligase chain reaction (LCR), this system facilitates highly sensitive and specific detection of single nucleotide polymorphisms (SNPs). Specifically, LCR precisely identifies single-base mutations, effectively transmitting biological information. The 3D DNA walker system is based on entropy-driven circuit cycling reaction technology. In this system, LCR products serve as the driving strands for the DNA walker, independently binding to track strands and walking legs immobilized on gold nanoparticles, forming a unique dual signal capture mechanism. Each track strand carries two signal chains, significantly enhancing signal amplification efficiency. Benefiting from this novel enzyme-free 3D DNA walker strategy, our biosensing system exhibits exceptional sensitivity to mutant targets (MT), detecting MT at concentrations as low as 30.3 aM and distinguishing heterozygous samples with a 0.01% mutation frequency. Furthermore, this system has been successfully applied to genotyping and mutation abundance assessment of genomes from fresh soybean leaves, demonstrating its vast potential for practical applications. In summary, this research pioneers a novel enzyme-free 3D DNA walker with dual capture and dual output capabilities, and develops an ultrasensitive genotyping tool. This provides strong technical support for the advancement of genetic research. © 2025 American Chemical Society.

关键词： Nucleotides

来源：评论

学校读者我要写书评

暂无评论

VFLMonitor: Defending One-Party Hijacking Attacks in Vertical Federated Learning

引用

IEEE Transactions on Information Forensics and Security 2025年 20卷 4828-4843页

作者： Xu, Xiangrui Zhao, Yiwen Han, Yufei Zhu, Yongsheng Han, Zhen Xu, Guangquan Wang, Bin Ji, Shouling Wang, Wei Qingdao University of Science and Technology School of Information Science and Technology Qingdao266061 China University of Southern California Thomas Lord Department of Computer Science Los AngelesCA90007 United States INRIA Brittany Rennes35042 France Institute of Computing Technologies China Academy of Railway Sciences Corporation Ltd. Beijing100081 China Beijing Jiaotong University Beijing Key Laboratory of Security and Privacy in Intelligent Transportation Beijing100044 China College of Intelligence and Computing Tianjin300072 China Network and Data Security Hangzhou310053 China Zhejiang University College of Computer Science and Technology Zhejiang 310058 China Xi’an Jiaotong University Ministry of Education Key Laboratory for Intelligent Networks and Network Security Xi’an710049 China

Vertical Federated Learning (VFL) is susceptible to various one-party hijacking attacks, such as Replay and Generation attacks, where a single malicious client can manipulate the model to produce attacker-specified results, thereby compromising its reliability in real-world deployments. In this paper, we first uncover the underlying mechanisms of these attacks and observe that successful attacks induce significant discrepancies in the embedding-label associations across different clients. We establish a theoretical framework demonstrating how these discrepancies can serve as reliable indicators for detecting hijacking attempts. Building upon this insight, we propose VFLMonitor, a robust defense mechanism that leverages these embedding-label discrepancies to detect and mitigate hijacking attacks. Specifically, VFLMonitor identifies suspicious queries by analyzing differences in label estimations from multiple clients and applies a majority voting rule to correct or filter out these malicious queries. Moreover, VFLMonitor introduces a novel regularization strategy during training to reduce intra-class variance in embeddings, thereby enhancing their discriminative power and improving defense effectiveness. Extensive experiments were conducted on 5 real-world datasets against 2 different attack types under 3 attack scenarios. The results demonstrate that VFLMonitor can effectively identify and exclude potential hijacked requests in all types of one-party hijacking attacks, while maintaining a meager false positive rate for legitimate queries. © 2005-2012 IEEE.

关键词： Training Servers Electronic mail Artificial intelligence Reliability theory Predictive models Federated learning Data models Computational modeling Collaboration

来源：评论

学校读者我要写书评

暂无评论

RobNODDI: Robust NODDI Parameter Estimation with Adaptive Sampling Under Continuous Representation 15th

RobNODDI: Robust NODDI Parameter Estimation with Adaptive S...

引用

15th International Workshop on Computational Diffusion MRI, CDMRI 2024, held in conjunction with 27th International Conference on Medical Image computing and Computer-Assisted Intervention, MICCAI 2024

作者： Xiao, Taohui Cheng, Jian Fan, Wenxin Yang, Jing Li, Cheng Dong, Enqing Wang, Shanshan School of Mechanical Electrical and Information Engineering Shandong University Weihai264209 China State Key Laboratory of Software Development Environment Beihang University Beijing China Key Laboratory of Data Science and Intelligent Computing Institute of International Innovation Beihang University Zhejiang Hangzhou China Paul C. Lauterbur Research Center for Biomedical Imaging Shenzhen Institute of Advanced Technology Chinese Academy of Sciences Guangdong Shenzhen China Peng Cheng Laboratory Guangdong Shenzhen China

ISBN: (纸本)9783031869198

Neurite Orientation Dispersion and Density Imaging (NODDI) is an important imaging technology used to evaluate the microstructure of brain tissue, which is of great significance for the discovery and treatment of various neurological diseases. Current deep learning-based methods perform parameter estimation through diffusion magnetic resonance imaging (dMRI) with a small number of diffusion gradients. These methods speed up parameter estimation and improve accuracy. However, the diffusion directions used by most existing deep learning models during testing needs to be strictly consistent with the diffusion directions during training. This results in poor generalization and robustness of deep learning models in dMRI parameter estimation. In this work, we first verify that the parameter estimation performance of current mainstream methods will significantly decrease when the testing diffusion directions and the training diffusion directions are inconsistent. A robust NODDI parameter estimation method with adaptive sampling under continuous representation (RobNODDI) is proposed. Furthermore, long short-term memory (LSTM) units and fully connected layers are selected to learn continuous representation signals. To this end, we use a total of 100 subjects to conduct experiments based on the Human Connectome Project (HCP) dataset, of which 60 are used for training, 20 are used for validation, and 20 are used for testing. The test results indicate that RobNODDI improves the generalization performance and robustness of the deep learning model, enhancing the stability and flexibility of deep learning NODDI parameter estimatimation applications. © The Author(s), under exclusive license to Springer Nature Switzerland AG 2025.

关键词： Deep learning

来源：评论

学校读者我要写书评

暂无评论

Fine-Grained Spatial-Temporal Framework for Engagement Prediction 2nd

Fine-Grained Spatial-Temporal Framework for Engagement Pred...

引用

2nd CSIG Conference on Emotional Intelligence, CEI 2024

作者： Wang, Shanshan Zheng, Yong Wang, keyang Zhang, Xingyi The Key Laboratory of Intelligent Computing and Signal Processing Ministry of Education Beijing China Institutes of Physical Science and Information Technology Anhui University Anhui China Zhejiang Dahua Technology Co. Ltd. Zhejiang China School of Computer Science and Technology Anhui University Anhui China

ISBN: (纸本)9789819650835

The engagement prediction task aims to identify the current level of involvement of individuals based on the information presented in video clips. In recent years, engagement prediction has attracted considerable attention due to its significance in real-life scenarios. We observed that an individual’s engagement status is often primarily manifested in the facial region. This observation prompted us to solve the engagement prediction problem by exploiting facial information to learn features related to engagement. In this paper, we propose a novel model for learning fine-grained features, called Fine-Grained Spatial-Temporal Framework (FGST). Specifically, we consider both spatial and temporal aspects. In the spatial domain, we propose an entropy-weighted method based on Fourier transform. By applying Fourier transform to video frames and introducing the concept of information entropy, we can segregate engagement-related information from engagement-unrelated information and mining difficult samples. This aids in enabling the model to learn features conducive to engagement prediction. In the temporal domain, we calculate the differences between adjacent frames and sum these differences to obtain regions of pixel variation between video frames, thereby generates saliency weight and improving model performance. Experiments on several public datasets demonstrate the proposed model can outperform the state-of-the-art methods. © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2025.

关键词： Engagement prediction Fourier transform Frame difference Information entropy

来源：评论

学校读者我要写书评

暂无评论

: heterogeneous hypergraph attention network with counterfactual learning for multimodal sentiment analysis

引用

Complex & intelligent Systems 2025年第4期11卷 1-16页

作者： Huang, Changqin Lin, Zhenheng Huang, Qionghao Huang, Xiaodi Jiang, Fan Chen, Jili Zhejiang Key Laboratory of Intelligent Education Technology and Application Zhejiang Normal University Jinhua China School of Education Zhejiang Normal University Jinhua China School of Computer Science and Technology Zhejiang Normal University Jinhua China School of Computing Mathematics and Engineering Charles Sturt University Albury Australia

Multimodal sentiment analysis (MSA) has garnered significant attention for its immense potential in human-computer interaction. While cross-modality attention mechanisms are widely used in MSA to capture inter-modality interactions, existing methods are limited to pairwise interactions between two modalities. Additionally, these methods can not utilize the causal relationship to guide attention learning, making them susceptible to bias information. To address these limitations, we introduce a novel method called Heterogeneous Hypergraph Attention Network with Counterfactual Learning $$(\text {H}^2\text {CAN}).$$ The method constructs a heterogeneous hypergraph based on sentiment expression characteristics and employs Heterogeneous Hypergraph Attention Networks (HHGAT) to capture interactions beyond pairwise constraints. Furthermore, it mitigates the effects of bias through a Counterfactual Intervention Task (CIT). Our model comprises two main branches: hypergraph fusion and counterfactual fusion. The former uses HHGAT to capture inter-modality interactions, while the latter constructs a counterfactual world using Gaussian distribution and additional weighting for the biased modality. The CIT leverages causal inference to maximize the prediction discrepancy between the two branches, guiding attention learning in the hypergraph fusion branch. We utilize unimodal labels to help the model adaptively identify the biased modality, thereby enhancing the handling of bias information. Experiments on three mainstream datasets demonstrate that $$\text {H}^2\text {CAN}$$ sets a new benchmark.

关键词：

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：