检索结果-内蒙古大学图书馆

Optimizing Human Pose Estimation Through Focused Human and Joint Regions

学校读者我要写书评

暂无评论

arXiv 2025年

作者： Jiao, Yingying Wang, Zhigang Liu, Zhenguang Fan, Shaojing Wu, Sifan Wu, Zheqi Xu, Zhuoyue College of Computer Science and Technology Jilin University China Key Laboratory of Symbolic Computation and Knowledge Engineering of Ministry of Education Jilin University China College of Computer Science and Technology Zhejiang Gongshang University China The State Key Laboratory of Blockchain and Data Security Zhejiang University China Institute of Blockchain and Data Security China School of Computing National University of Singapore Singapore

Human pose estimation has given rise to a broad spectrum of novel and compelling applications, including action recognition, sports analysis, as well as surveillance. However, accurate video pose estimation remains an open challenge. One aspect that has been overlooked so far is that existing methods learn motion clues from all pixels rather than focusing on the target human body, making them easily misled and disrupted by unimportant information such as background changes or movements of other people. Additionally, while the current Transformer-based pose estimation methods has demonstrated impressive performance with global modeling, they struggle with local context perception and precise positional identification. In this paper, we try to tackle these challenges from three aspects: (1) We propose a bilayer Human-keypoint Mask module that performs coarse-to-fine visual token refinement, which gradually zooms in on the target human body and keypoints while masking out unimportant figure regions. (2) We further introduce a novel deformable cross attention mechanism and a bidirectional separation strategy to adaptively aggregate spatial and temporal motion clues from constrained surrounding contexts. (3) We mathematically formulate the deformable cross attention, constraining that the model focuses solely on the regions centered at the target person body. Empirically, our method achieves state-of-the-art performance on three large-scale benchmark datasets. A remarkable highlight is that our method achieves an 84.8 mean Average Precision (mAP) on the challenging wrist joint, which significantly outperforms the 81.5 mAP achieved by the current state-of-the-art method on the PoseTrack2017 dataset. Copyright © 2025, The Authors. All rights reserved.

关键词： Benchmarking

SpatioTemporal Learning for Human Pose Estimation in Sparsely-Labeled Videos

学校读者我要写书评

暂无评论

arXiv 2025年

作者： Jiao, Yingying Wang, Zhigang Wu, Sifan Fan, Shaojing Liu, Zhenguang Xu, Zhuoyue Wu, Zheqi College of Computer Science and Technology Jilin University China Key Laboratory of Symbolic Computation and Knowledge Engineering of Ministry of Education Jilin University China College of Computer Science and Technology Zhejiang Gongshang University China School of Computing National University of Singapore Singapore The State Key Laboratory of Blockchain and Data Security Zhejiang University China Institute of Blockchain and Data Security China

Human pose estimation in videos remains a challenge, largely due to the reliance on extensive manual annotation of large datasets, which is expensive and labor-intensive. Furthermore, existing approaches often struggle to capture long-range temporal dependencies and overlook the complementary relationship between temporal pose heatmaps and visual features. To address these limitations, we introduce STDPose, a novel framework that enhances human pose estimation by learning spatiotemporal dynamics in sparsely-labeled videos. STDPose incorporates two key innovations: 1) A novel Dynamic-Aware Mask to capture long-range motion context, allowing for a nuanced understanding of pose changes. 2) A system for encoding and aggregating spatiotemporal representations and motion dynamics to effectively model spatiotemporal relationships, improving the accuracy and robustness of pose estimation. STDPose establishes a new performance benchmark for both video pose propagation (i.e., propagating pose annotations from labeled frames to unlabeled frames) and pose estimation tasks, across three large-scale evaluation datasets. Additionally, utilizing pseudo-labels generated by pose propagation, STDPose achieves competitive performance with only 26.7% labeled data. Copyright © 2025, The Authors. All rights reserved.

关键词： Large datasets

A Lightweight Authentication Protocol Based on Pid-Ga and Ecc for Mobile Edge Computing

学校读者我要写书评

暂无评论

SSRN

SSRN 2022年

作者： Zhang, Linghui Zhou, Changli Quan, Hanyu Wen, Jingliang College of Computer Science and Technology HuaQiao University Xiamen361021 China Xiamen Key Laboratory of Data Security and Blockchain Technology HuaQiao University Xiamen361021 China

The existing cloud computing network model is no longer applicable to the fast-growing edge terminal environment, and decentralized network model has gradually become a hot topic of concern. However, in the edge computing environment, the existing authentication protocols, on the hand, they lack high security, and on the other hand always use expensive bilinear pairing operations, which do not well meet the network requirements of low latency and high security in edge computing. To this end, we propose a lightweight authentication protocol based on ECC and improve a security algorithm called PID-GA. This algorithm guarantees more user anonymity and privacy security than the traditional method of generating pseudo-identities. In addition, we perform formal and informal security analysis to prove the security of the protocol;and we demonstrate that our scheme outperforms other schemes in terms of both computational and communication overheads through performance comparisons. © 2022, The Authors. All rights reserved.

关键词： Authentication

A Novel Lexicon Enhanced Named Entity Recognition Model for Network security Domain

学校读者我要写书评

暂无评论

A Novel Lexicon Enhanced Named Entity Recognition Model for ...

2024 China Automation Congress, CAC 2024

作者： Jin, Zhiyang Fu, Yusun Huan, Yuehui Xue, Jinliang Wang, Chengguang Chen, Yuanfang Ningbo Artificial Intelligence Institute Shanghai Jiao Tong University Ningbo China School of Electronic Information and Electrical Engineering Shanghai Jiao Tong University Shanghai China Zhejiang Guoli Security Technology Co. Ltd Ningbo China Ningbo Industrial Internet Institute Ningbo China State Key Laboratory of Blockchain and Data Security Zhejiang University Hangzhou China

ISBN: (纸本)9798350368604

Named Entity Recognition (NER) technology serves as the foundation for solving many Natural Language Processing(NLP) problems, particularly in the domain of building knowledge graphs, where an effective NER model is crucial for creating efficient knowledge representations. However, when constructing knowledge graphs in the Industrial Internet security(IIS) domain, security-related incidents, as significant entities, pose challenges for existing NER models. In this paper, we propose security Lexicon Enhanced Bidirectional Encoder Representations from Transformers(SEBERT). SEBERT enhances semantic processing by integrating security lexicon information into one transformer layer of BERT, assigning vocabulary to each character in input sentences. Then we use a Residual-Knowledge-Arnold Network(RKAN) to optimize the semantic features learned from SEBERT, utilizing the properties of residual networks to learn identity mapping features without significantly increasing computational complexity. Finally, the Conditional Random Field (CRF) is used to learn the sequence information and dependencies of adjacent labels. Experimental results demonstrate that SEBERT-RKAN-CRF model achieves superior performance in recognizing entities in the IIS domain, with improvements of 1.99%, 3.34%, and 2.66% in accuracy rate, recall rate, and F1-score compared to the baseline model. © 2024 IEEE.

关键词： Semantics

Safeguarding LLM Embeddings in End-Cloud Collaboration via Entropy-Driven Perturbation

学校读者我要写书评

暂无评论

arXiv 2025年

作者： Jin, Shuaifan Pang, Xiaoyi Wang, Zhibo Wang, He Du, Jiacheng Hu, Jiahui Ren, Kui The State Key Laboratory of Blockchain and Data Security Zhejiang University China School of Cyber Science and Technology Zhejiang University China Hong Kong University of Science and Technology Hong Kong

Recent studies improve on-device language model (LM) inference through end-cloud collaboration, where the end device retrieves useful information from cloud databases to enhance local processing, known as Retrieval-Augmented Generation (RAG). Typically, to retrieve information from the cloud while safeguarding privacy, the end device transforms original data into embeddings with a local embedding model. However, the recently emerging Embedding Inversion Attacks (EIAs) can still recover the original data from text embeddings (e.g., training a recovery model to map embeddings back to original texts), posing a significant threat to user privacy. To address this risk, we propose EntroGuard, an entropy-driven perturbation-based embedding privacy protection method, which can protect the privacy of text embeddings while maintaining retrieval accuracy during the end-cloud collaboration. Specifically, to defeat various EIAs, we perturb the embeddings to increase the entropy of the recovered text in the common structure of recovery models, thus steering the embeddings toward meaningless texts rather than original sensitive texts during the recovery process. To maintain retrieval performance in the cloud, we constrain the perturbations within a bound, applying the strategy of reducing them where redundant and increasing them where sparse. Moreover, EntroGuard can be directly integrated into end devices without requiring any modifications to the embedding model. Extensive experimental results demonstrate that EntroGuard can reduce the risk of privacy leakage by up to 8 times at most with negligible loss of retrieval performance compared to existing privacy-preserving methods. Copyright © 2025, The Authors. All rights reserved.

关键词： Differential privacy

Towards Fair Federated Learning via Unbiased Feature Aggregation

学校读者我要写书评

暂无评论

IEEE Transactions on Dependable and Secure Computing 2025年

作者： He, Zeqing Wang, Zhibo Dong, Xiaowei Sun, Peng Ren, Ju Ren, Kui Zhejiang University State Key Laboratory of Blockchain and Data Security China Institute of Blockchain and Data Security Hangzhou310027 China Wuhan University School of Cyber Science and Engineering Wuhan430072 China Hunan University College of Computer Science and Electronic Engineering Changsha410082 China Tsinghua University Department of Computer Science and Technology Beijing100084 China Zhongguancun Laboratory Beijing100094 China

Federated learning (FL) is a distributed machine learning framework that enables multiple clients to collaboratively train models without raw data exchange. Prior studies on FL mainly focus on optimizing learning performance, enhancing privacy preservation, and improving attack resilience. However, little work studies how to mitigate the unfairness of federated trained models while unfair models would make discriminatory decisions toward certain groups or populations (e.g., favoring males over females), leading to serious ethical concerns. Thus, it is crucial to mitigate model unfairness in FL, yet challenging as this requires centralized access to each data point's fairness-sensitive information (e.g., race, gender), which is prohibited in FL. In this work, we propose a novel fair FL framework FedUFA, where the server can aggregate clients' learned knowledge in an unbiased manner, to obtain fair and high-usability federated trained models. Specifically, to unearth the bias in clients' local data and account for potentially heterogeneous local models, we propose a knowledge distillation-based FL scheme, where clients' knowledge of learned features on a public dataset is amalgamated to the server for aggregation. We train an unbiased feature mapper at the server to remove fairness-sensitive latent features and extract fair representations from clients' submitted raw features. In particular, we design an adversarial training method to train the mapper, which involves a predictor aiming to maximize the prediction accuracy on the FL task and a discriminator intending to help identify fairness-sensitive features. Extensive experiments on real-world datasets demonstrate the effectiveness of FedUFA. © 2025 IEEE. All rights reserved,

关键词： Federated learning

Uncertainty-aware complementary label queries for active learning

学校读者我要写书评

暂无评论

Frontiers of Information technology & Electronic Engineering 2023年第10期24卷 1497-1503页

作者： Shengyuan LIU Ke CHEN Tianlei HU Yunqing MAO Key Lab of Intelligent Computing Based Big Data of Zhejiang Province Zhejiang UniversityHangzhou 310027China State Key Laboratory of Blockchain and Data Security Zhejiang UniversityHangzhou 310027China City Cloud Technology(China)Co. Ltd.Hangzhou 310000China

Many active learning methods assume that a learner can simply ask for the full annotations of some training data from *** methods mainly try to cut the annotation costs by minimizing the number of annotation ***,annotating instances exactly in many realworld classification tasks is still *** reduce the cost of a single annotation action,we try to tackle a novel active learning setting,named active learning with complementary labels(ALCL).ALCL learners ask only yes/no questions in some *** receiving answers from annotators,ALCL learners obtain a few supervised instances and more training instances with complementary labels,which specify only one of the classes to which the pattern does not *** are two challenging issues in ALCL:one is how to sample instances to be queried,and the other is how to learn from these complementary labels and ordinary accurate *** the first issue,we propose an uncertainty-based sampling strategy under this novel setup.

关键词： complementary instance ordinary

DSGNN: A Dual-View Supergrid-Aware Graph Neural Network for Regional Air Quality Estimation

学校读者我要写书评

暂无评论

arXiv 2024年

作者： Zhang, Xin Chen, Ling Tang, Xing Shi, Hongyu The State Key Laboratory of Blockchain and Data Security Zhejiang University Hangzhou310027 China The College of Computer Science and Technology Zhejiang University Hangzhou310027 China

Air quality estimation can provide air quality for target regions without air quality stations, which is useful for the public. Existing air quality estimation methods divide the study area into disjointed grid regions, and apply 2D convolution to model the spatial dependencies of adjacent grid regions based on the first law of geography, failing to model the spatial dependencies of distant grid regions. To this end, we propose a Dual-view Supergrid-aware Graph Neural Network (DSGNN) for regional air quality estimation, which can model the spatial dependencies of distant grid regions from dual views (i.e., satellite-derived aerosol optical depth (AOD) and meteorology). Specifically, images are utilized to represent the regional data (i.e., AOD data and meteorology data). The dual-view supergrid learning module is introduced to generate supergrids in a parameterized way. Based on the dual-view supergrids, the dual-view implicit correlation encoding module is introduced to learn the correlations between pairwise supergrids. In addition, the dual-view message passing network is introduced to implement the information interaction on the supergrid graphs and images. Extensive experiments on two real-world datasets demonstrate that DSGNN achieves the state-of-the-art performances on the air quality estimation task, outperforming the best baseline by an average of 19.64% in MAE. Copyright © 2024, The Authors. All rights reserved.

关键词： Air quality

D-PAD: Deep-Shallow Multi-Frequency Patterns Disentangling for Time Series Forecasting

学校读者我要写书评

暂无评论

arXiv 2024年

作者： Yuan, Xiaobing Chen, Ling The State Key Laboratory of Blockchain and Data Security Zhejiang University Hangzhou310027 China The College of Computer Science and Technology Zhejiang University Hangzhou310027 China

In time series forecasting, effectively disentangling intricate temporal patterns is crucial. While recent works endeavor to combine decomposition techniques with deep learning, multiple frequencies may still be mixed in the decomposed components, e.g., trend and seasonal. Furthermore, frequency domain analysis methods, e.g., Fourier and wavelet transforms, have limitations in resolution in the time domain and adaptability. In this paper, we propose D-PAD, a Deep-shallow multi-frequency PAtterns Disentangling neural network for time series forecasting. Specifically, a multi-component decomposing (MCD) block is introduced to decompose the series into components with different frequency ranges, corresponding to the "shallow" aspect. A decomposition-reconstruction-decomposition (D-R-D) module is proposed to progressively extract the information of frequencies mixed in the components, corresponding to the "deep" aspect. After that, an interaction and fusion (IF) module is used to further analyze the components. Extensive experiments on seven real-world datasets demonstrate that D-PAD achieves the state-of-the-art performance, outperforming the best baseline by an average of 9.48% and 7.15% in MSE and MAE, respectively. Copyright © 2024, The Authors. All rights reserved.

关键词： Frequency domain analysis