检索结果-内蒙古大学图书馆

Conference on computer Vision and Pattern Recognition (CVPR)

作者： Xiaogeng Liu Minghui Li Haoyu Wang Shengshan Hu Dengpan Ye Hai Jin Libing Wu Chaowei Xiao School of Cyber Science and Engineering Huazhong University of Science and Technology National Engineering Research Center for Big Data Technology and System Services Computing Technology and System Lab Hubei Key Laboratory of Distributed System Security Hubei Engineering Research Center on Big Data Security School of Software Engineering Huazhong University of Science and Technology School of Cyber Science and Engineering Wuhan University School of Computer Science and Technology Huazhong University of Science and Technology Cluster and Grid Computing Lab Arizona State University

Deep neural networks are proven to be vulnerable to backdoor attacks. Detecting the trigger samples during the inference stage, i.e., the test-time trigger sample detection, can prevent the backdoor from being triggered. However, existing detection methods often require the defenders to have high accessibility to victim models, extra clean data, or knowledge about the appearance of backdoor triggers, limiting their practicality. In this paper, we propose the test-time corruption robustness consistency evaluation (TeCo) 1 1 https://***/CGCL-codes/TeCo, a novel test-time trigger sample detection method that only needs the hard-label outputs of the victim models without any extra information. Our journey begins with the intriguing observation that the backdoor-infected models have similar performance across different image corruptions for the clean images, but perform discrepantly for the trigger samples. Based on this phenomenon, we design TeCo to evaluate test-time robustness consistency by calculating the deviation of severity that leads to predictions' transition across different corruptions. Extensive experiments demonstrate that compared with state-of-the-art defenses, which even require either certain information about the trigger types or accessibility of clean data, TeCo outperforms them on different backdoor attacks, datasets, and model architectures, enjoying a higher AUROC by 10% and 5 times of stability.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Not All Pixels Are Equal: Learning Pixel Hardness for Semantic Segmentation

arXiv

引用

arXiv 2023年

作者： Xiao, Xin Zhou, Daiguo Hu, Jiagao Hu, Yi Xu, Yongchao National Engineering Research Center for Multimedia Software Institute of Artificial Intelligence School of Computer Science Hubei Key Laboratory of Multimedia and Network Communication Engineering Wuhan University Wuhan430070 China Xiaomi AI Lab Wuhan430070 China Hubei Medical Devices Quality Supervision and Test Institute Hubei Center for AI Medical Devices Supercomputing Wuhan430070 China

Semantic segmentation has recently witnessed great progress. Despite the impressive overall results, the segmentation performance in some hard areas (e.g., small objects or thin parts) is still not promising. A straightforward solution is hard sample mining, which is widely used in object detection. Yet, most existing hard pixel mining strategies for semantic segmentation often rely on pixel’s loss value, which tends to decrease during training. Intuitively, the pixel hardness for segmentation mainly depends on image structure and is expected to be stable. In this paper, we propose to learn pixel hardness for semantic segmentation, leveraging hardness information contained in global and historical loss values. More precisely, we add a gradient-independent branch for learning a hardness level (HL) map by maximizing hardness-weighted segmentation loss, which is minimized for the segmentation head. This encourages large hardness values in difficult areas, leading to appropriate and stable HL map. Despite its simplicity, the proposed method can be applied to most segmentation methods with no and marginal extra cost during inference and training, respectively. Without bells and whistles, the proposed method achieves consistent/significant improvement (1.37% mIoU on average) over most popular semantic segmentation methods on Cityscapes dataset, and demonstrates good generalization ability across domains. The source codes are available at https://***/Menoly-xin/Hardness-Level-Learning. Copyright © 2023, The Authors. All rights reserved.

关键词： Semantic Segmentation

来源：评论

学校读者我要写书评

暂无评论

Manifold Projection for Adversarial Defense on Face Recognition 1

引用

16th European Conference on computer Vision, ECCV 2020

作者： Zhou, Jianli Liang, Chao Chen, Jun National Engineering Research Center for Multimedia Software School of Computer Science Wuhan University Wuhan China Key Laboratory of Multimedia and Network Communication Engineering BeijingHubei Province China

ISBN: (数字)9783030585778

ISBN: (纸本)9783030585761

Although deep convolutional neural network based face recognition system has achieved remarkable success, it is susceptible to adversarial images: carefully constructed imperceptible perturbations can easily mislead deep neural networks. A recent study has shown that in addition to regular off-manifold adversarial images, there are also adversarial images on the manifold. In this paper, we propose Adversarial Variational AutoEncoder (A-VAE), a novel framework to tackle both types of attacks. We hypothesize that both off-manifold and on-manifold attacks move the image away from the high probability region of image manifold. We utilize variational autoencoder (VAE) to estimate the lower bound of the log-likelihood of image and explore to project the input images back into the high probability regions of image manifold again. At inference time, our model synthesizes multiple similar realizations of a given image by random sampling, then the nearest neighbor of the given image is selected as the final input of the face recognition model. As a preprocessing operation, our method is attack-agnostic and can adapt to a wide range of resolutions. The experimental results on LFW demonstrate that our method achieves state-of-the-art defense success rate against conventional off-manifold attacks such as FGSM, PGD, and C&W under both grey-box and white-box settings, and even on-manifold attack. © 2020, Springer Nature Switzerland AG.

关键词： Face recognition

来源：评论

学校读者我要写书评

暂无评论

Learning Dynamic Behavior Patterns for Fraud Detection

Learning Dynamic Behavior Patterns for Fraud Detection

引用

IEEE International Conference on Big Data and Cloud Computing (BdCloud)

作者： Zijun Huang Junhang Wu Lingfei Ren Ruimin Hu Dengshi Li National Engineering Research Center for Multimedia Software School of Computer Science Wuhan University Wuhan China Hubei Key Laboratory of Multimedia and Network Communication Engineering Wuhan University Wuhan China School of Cyber Engineering Xidian University Xi'an China School of Artificial Intelligence Jianghan University Wuhan China

Telecom fraud detection is considered a needle in a haystack task due to the fact that fraudsters are usually hidden among a large number of normal users. In fact, a fraudster will usually act in a continuous manner over a period of time to achieve fraudulent purposes. At the same time, fraudsters are constantly looking for new fraudulent targets. Therefore, we consider a behavioral pattern-based fraudster identification system called HADyn-LSTM for telecommunication network fraud detection. Motivated by the breakthroughs in deep learning (DL), we make an effort to judge whether a user is a fraudster is based on the user's behavior sequence in the social network, which means that we study the difference between normal users and fraudsters through dynamic behavior patterns rather than static statistical attributes. Particularly, Based on sequential neural networks and attention mechanisms, we design an end-to-end system that learns and extracts the dynamic patterns of user behavior sequences. Furthermore, We tested our system on three datasets of different sizes. The test results show that the HADyn-LSTM system performs better than the static attribute-based method.

关键词： Social networking (online) Neural networks Telephone sets Behavioral sciences Fraud Telecommunications Safety

来源：评论

学校读者我要写书评

暂无评论

PPO-Based Vehicle Control for Ramp Merging Scheme Assisted by Enhanced C-V2X

arXiv

引用

arXiv 2025年

作者： Wu, Qiong Ji, Maoxin Fan, Pingyi Wang, Kezhi Cheng, Nan Chen, Wen Letaief, Khaled B. The School of Internet of Things Engineering Jiangnan University Wuxi214122 China The Department of Electronic Engineering State Key laboratory of Space Network and Communications Beijing National Research Center for Information Science and Technology Tsinghua University Beijing100084 China The Department of Computer Science Brunel University Middlesex LondonUB8 3PH United Kingdom The State Key Lab. of ISN School of Telecommunications Engineering Xidian University Xi’an710071 China The Department of Electronic Engineering Shanghai Jiao Tong University Shanghai200240 China Hong Kong

On-ramp merging presents a critical challenge in autonomous driving, as vehicles from merging lanes need to dynamically adjust their positions and speeds while monitoring traffic on the main road to prevent collisions. To address this challenge, we propose a novel merging control scheme based on reinforcement learning, which integrates lateral control mechanisms. This approach ensures the smooth integration of vehicles from the merging lane onto the main road, optimizing both fuel efficiency and passenger comfort. Furthermore, we recognize the impact of vehicle-to-vehicle (V2V) communication on control strategies and introduce an enhanced protocol leveraging Cellular Vehicle-to-Everything (C-V2X) Mode 4. This protocol aims to reduce the Age of Information (AoI) and improve communication reliability. In our simulations, we employ two AoI-based metrics to rigorously assess the protocol's effectiveness in autonomous driving scenarios. By combining the NS3 network simulator with Python, we simulate V2V communication and vehicle control simultaneously. The results demonstrate that the enhanced C-V2X Mode 4 outperforms the standard version, while the proposed control scheme ensures safe and reliable vehicle operation during on-ramp merging. Copyright © 2025, The Authors. All rights reserved.

关键词： Problem oriented languages

来源：评论

学校读者我要写书评

暂无评论

Asynchronous Tag-based Physical-Layer Authentication in Wireless Communications

引用

IEEE Transactions on Wireless Communications 2025年

作者： Tan, Haijun Xu, Yuntao Du, Jiewei Xie, Ning Li, Hongbin Peng Cheng Laboratory Control and Network Foundation Laboratory Frontier Research Center Shenzhen518107 China Shenzhen University State Key Laboratory of Radio Frequency Heterogeneous Integration The Guangdong Key Laboratory of Intelligent Information Processing College of Electronics and Information Engineering Shenzhen518060 China Beijing Jiaotong University School of Electronic and Information Engineering Beijing100044 China National Innovation Center of High Speed Train Qingdao266111 China Stevens Institute of Technology Department of Electrical and Computer Engineering HobokenNJ07030 United States

Tag-based physical-layer authentication (PLA) has gained significant research interest due to its high security and low complexity compared to traditional upper-layer authentication mechanisms. However, conventional tag-based PLA schemes often require strict synchronization between the tag and the source message before transmission, which may limit their practicality and effectiveness. To address this limitation, we propose an asynchronous tag-based PLA scheme, where the tag and the source message are transmitted asynchronously with an intentional time delay introduced at the transmitter. The proposed scheme enhances both compatibility and security without compromising robustness. We provide a theoretical analysis of the proposed scheme, and derive closed-form expressions to characterize its performance. Furthermore, theoretical comparisons with existing methods yield insightful conclusions regarding the advantages of the proposed approach. The scheme is implemented and extensively evaluated through simulations, where the theoretical results exhibit strong agreement with the simulation outcomes, thereby validating its effectiveness. © 2025 IEEE.

关键词： asynchronous closed form intentional timing delay Physical-layer authentication tag-based

来源：评论

学校读者我要写书评

暂无评论

Soft Tissue Deformation Model Based on Marquardt Algorithm and Enrichment Function

引用

computer Modeling in engineering & Sciences 2020年第9期124卷 1131-1147页

作者： Xiaorui Zhang Xuefeng Yu Wei Sun Aiguo Song Jiangsu Engineering Center of Network Monitoring Engineering Research Center of Digital ForensicsMinistry of EducationSchool of Computer and SoftwareNanjing University of Information Science&TechnologyNanjing210044China Jiangsu Collaborative Innovation Center of Atmospheric Environment and Equipment Technology Nanjing University of Information Science&TechnologyNanjing210044China State Key Laboratory of Bioelectronics Jiangsu Key Lab of Remote Measurement and ControlSchool of Instrument Science and EngineeringSoutheast UniversityNanjing210096China

In order to solve the problem of high computing cost and low simulation accuracy caused by discontinuity of incision in traditional meshless model,this paper proposes a soft tissue deformation model based on the Marquardt algorithm and enrichment *** model is based on the element-free Galerkin method,in which Kelvin viscoelastic model and adjustment function are *** algorithm is applied to fit the relation between force and displacement caused by surface deformation,and the enrichment function is applied to deal with the discontinuity in the meshless *** verify the validity of the model,the Sensable Phantom Omni force tactile interactive device is used to simulate the deformations of stomach and *** results show that the proposed model improves the real-time performance and accuracy of soft tissue deformation simulation,which provides a new perspective for the application of the meshless method in virtual surgery.

关键词： Virtual surgery meshless model Marquardt algorithm enrichment function soft tissue simulation

来源：评论

学校读者我要写书评

暂无评论

On the parity-check matrix of generalized concatenated code

引用

Science China(Information Sciences) 2020年第4期63卷 233-235页

作者： Sha SHI Yang LIU Junzhi YAN Jingliang GAO Yunjiang WANG Engineering Research Center of Molecular and Neuro Imaging Ministry of Education School of Life Science and Technology Xidian University National Computer Network Emergency Response Technical Team/Coordination Center of China China Mobile Research Institute State Key Laboratory of Integrated Services Networks Xidian University

Dear editor,Single-level concatenation using a nonbinary code as outer code and a binary code as inner code is widely used in communication and digital data storage systems to achieve high reliability with reduced decoding complexity [1]. Moreover, concatenation can be multilevel, where multiple outer codes are concatenated with multiple inner codes

关键词： On the parity-check matrix of generalized concatenated code

来源：评论

学校读者我要写书评

暂无评论

Hyper-relationship Learning network for Scene Graph Generation

arXiv

引用

arXiv 2022年

作者： Zhan, Yibing Chen, Zhi Yu, Jun Yu, Bao Sheng Luo, Yong Tao, Dacheng JD Explore Academy China Hangzhou Dianzi University China The University of Sydney Australia National Engineering Research Center for Multimedia Software Institute of Artificial Intelligence School of Computer Science Hubei Key Laboratory of Multimedia and Network Communication Engineering Wuhan University China

Generating informative scene graphs from images requires integrating and reasoning from various graph components, i.e., objects and relationships. However, current scene graph generation (SGG) methods, including the unbiased SGG methods, still struggle to predict informative relationships due to the lack of 1) high-level inference such as transitive inference between relationships and 2) efficient mechanisms that can incorporate all interactions of graph components. To address the issues mentioned above, we devise a hyper-relationship learning network, termed HLN, for SGG. Specifically, the proposed HLN stems from hypergraphs and two graph attention networks (GATs) are designed to infer relationships: 1) the object-relationship GAT or OR-GAT to explore interactions between objects and relationships, and 2) the hyper-relationship GAT or HR-GAT to integrate transitive inference of hyper-relationships, i.e., the sequential relationships between three objects for transitive reasoning. As a result, HLN significantly improves the performance of scene graph generation by integrating and reasoning from object interactions, relationship interactions, and transitive inference of hyper-relationships. We evaluate HLN on the most popular SGG dataset, i.e., the Visual Genome dataset, and the experimental results demonstrate its great superiority over recent state-of-the-art methods. For example, the proposed HLN improves the recall per relationship from 11.3% to 13.1%, and maintains the recall per image from 19.8% to 34.9%. We will release the source code and pretrained models on GitHub. Copyright © 2022, The Authors. All rights reserved.

关键词： Machine learning

来源：评论

学校读者我要写书评

暂无评论

Prototype-Wise Self-Knowledge Distillation for Few-Shot Segmentation

SSRN

引用

SSRN 2023年

作者： Chen, Yadang Xu, Xinyu Wei, Chenchen Yang, Zhi-Xin School of Computer and Software Nanjing University of Information Science and Technology Nanjing210044 China Engineering Research Center of Digital Forensics Ministry of Education Nanjing University of Information Science and Technology Nanjing210044 China The State Key Laboratory of Internet of Things for Smart City Department of Electromechanical Engineering University of Macau 999078 China

Few-shot segmentation was proposed to obtain segmentation results for a image with an unseen class by referring to a few labeled samples. However, due to the limited number of samples, many few-shot segmentation models suffer from poor generalization. Prototypical network-based few-shot segmentation still has issues with spatial inconsistency and prototype bias. Since the target class has different appearance in each image, some specific features in the prototypes generated from the support image and its mask do not accurately reflect the generalized features of the target class. To address the support prototype consistency issue, we put forward two modules: Data Augmentation Self-knowledge Distillation (DASKD) and Prototype-wise Regularization (PWR). The DASKD module focuses on enhancing spatial consistency by using data augmentation and self-knowledge distillation. Self-knowledge distillation helps the model acquire generalized features of the target class and learn hidden knowledge from the support images. The PWR module focuses on obtaining a more representative support prototype by conducting prototype-level loss to obtain support prototypes closer to the category center. Broad evaluation experiments on PASCAL-5i and COCO-20i demonstrate that our model outperforms the prior works on few-shot segmentation. Our approach surpasses the state of the art by 7.5% in PASCAL-5i and 4.2% in COCO-20i. © 2023, The Authors. All rights reserved.

关键词： Distillation

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：