检索结果-内蒙古大学图书馆

arXiv 2024年

作者： Tian, Chunwei Zheng, Menghua Li, Bo Zhang, Yanning Zhang, Shichao Zhang, David School of Software Northwestern Polytechnical University Xi’an710129 China National Engineering Laboratory for Integrated Aero-Space-Ground-Ocean Big Data Application Technology Xi’an710129 China School of Electronics and Information Northwestern Polytechnical University Xi’an710129 China School of Computer Science Northwestern Polytechnical University National Engineering Laboratory for Integrated AeroSpace-Ground-Ocean Big Data Application Technology Xi’an710129 China Guangxi Key Lab of Multisource Information Mining & Security College of Computer Science & Engineering Guangxi Normal University Guilin541004 China Shenzhen518172 China Shenzhen Institute of Artificial Intelligence and Robotics for Society Shenzhen518172 China

Popular methods usually use a degradation model in a supervised way to learn a watermark removal model. However, it is true that reference images are difficult to obtain in the real world, as well as collected images by cameras suffer from noise. To overcome these drawbacks, we propose a perceptive self-supervised learning network for noisy image watermark removal (PSLNet) in this paper. PSLNet depends on a parallel network to remove noise and watermarks. The upper network uses task decomposition ideas to remove noise and watermarks in sequence. The lower network utilizes the degradation model idea to simultaneously remove noise and watermarks. Specifically, mentioned paired watermark images are obtained in a self-supervised way, and paired noisy images (i.e., noisy and reference images) are obtained in a supervised way. To enhance the clarity of obtained images, interacting two sub-networks and fusing obtained clean images are used to improve the effects of image watermark removal in terms of structural information and pixel enhancement. Taking into texture information account, a mixed loss uses obtained images and features to achieve a robust model of noisy image watermark removal. Comprehensive experiments show that our proposed method is very effective in comparison with popular convolutional neural networks (CNNs) for noisy image watermark removal. Codes can be obtained at https://***/hellloxiaotian/PSLNet. Copyright © 2024, The Authors. All rights reserved.

关键词： Image denoising

来源：评论

学校读者我要写书评

暂无评论

Multi-View Mutual Learning Network for Multimodal Fake News Detection

SSRN

引用

SSRN 2023年

作者： Cui, Wei Zhang, Xuerui Shang, Mingsheng College of Computer Science and Technology Chongqing University of Posts and Telecommunications Chongqing400065 China Chongqing Key Laboratory of Big Data and Intelligent Computing Chongqing Institute of Green and Intelligent Technology Chinese Academy of Sciences Chongqing400714 China School of Electronic Information and Communication Engineering Chongqing Aerospace Polytechnic Chongqing400021 China

Multimodal fake news is more deceptive than unimodal content and often has adverse social and economic impacts. However, most existing methods learn modal features from a single perspective, without considering simultaneously learning and sharing knowledge across modalities from different perspectives. Our work presents a novel Multi-View Mutual Learning (MVML) network for multimodal fake news detection, which explores the semantic relationship between words and scenes, as well as words and objects, and investigates the semantic connection that exists between global concepts and local objects from multiple perspectives. We first construct text-scenes and text-objects graphs respectively, and perform intra-graph inference to obtain multi-view features. Then, the multi-view fusion layer based on the cross-view attention mechanism interactively models the cross-view dependencies between the image and text. The extracted features before feeding into the final classifier are further processed via a feed-forward attention module, which adaptively reweights and aggregates the features for redundancy reduction. In addition, to exploit the respective advantages of two multi-view classifiers, we propose a mutual learning mechanism that allows them to perform knowledge distillation and align the learning targets. The proposed MVML is thoroughly assessed on four publicly available benchmark datasets, and the findings demonstrate that it outperforms the existing standard approaches. © 2023, The Authors. All rights reserved.

关键词： Semantics

来源：评论

学校读者我要写书评

暂无评论

FaceBench: A Multi-View Multi-Level Facial Attribute VQA dataset for Benchmarking Face Perception MLLMs

arXiv

引用

arXiv 2025年

作者： Wang, Xiaoqin Ma, Xusen Hou, Xianxu Ding, Meidan Li, Yudong Chen, Junliang Chen, Wenting Peng, Xiaoyang Shen, Linlin Computer Vision Institute College of Computer Science and Software Engineering Shenzhen University China National Engineering Laboratory for Big Data System Computing Technology Shenzhen University China Guangdong Provincial Key Laboratory of Intelligent Information Processing China AIAC Xi’an Jiaotong-Liverpool University China Tsinghua University China The Hong Kong Polytechnic University Hong Kong City University of Hong Kong Hong Kong Sun Yat-sen University China

Multimodal large language models (MLLMs) have demonstrated remarkable capabilities in various tasks. However, effectively evaluating these MLLMs on face perception remains largely unexplored. To address this gap, we introduce FaceBench, a dataset featuring hierarchical multi-view and multi-level attributes specifically designed to assess the comprehensive face perception abilities of MLLMs. Initially, we construct a hierarchical facial attribute structure, which encompasses five views with up to three levels of attributes, totaling over 210 attributes and 700 attribute values. Based on the structure, the proposed FaceBench consists of 49,919 visual question-answering (VQA) pairs for evaluation and 23,841 pairs for fine-tuning. Moreover, we further develop a robust face perception MLLM baseline, Face-LLaVA, by training with our proposed face VQA data. Extensive experiments on various mainstream MLLMs and Face-LLaVA are conducted to test their face perception ability, with results also compared against human performance. The results reveal that, the existing MLLMs are far from satisfactory in understanding the fine-grained facial attributes, while our Face-LLaVA significantly outperforms existing open-source models with a small amount of training data and is comparable to commercial ones like GPT-4o and Gemini. The dataset will be released at https://***/CVI-SZU/FaceBench. Copyright © 2025, The Authors. All rights reserved.

关键词：

来源：评论

学校读者我要写书评

暂无评论

2-D GCANet Applied to Denoise 1-D EEG Signals in Online Remote Teaching Scenarios

Journal of Network Intelligence

引用

Journal of Network Intelligence 2023年第4期8卷 1289-1302页

作者： Huang, Ting-Qing Wang, Chuan-Sheng Chen, Zhao-Qi Zhang, Fu-Quan Meng, Xiang-Long Grau, Antoni Chen, Yang Huang, Jing-Wei College of Computer and Control Engineering Minjiang University Fuzhou University Town No. 200 Xiyuangong Road Fuzhou China Department of Automatic Control Technical Polytechnic University of Catalonia Autonomous Region of Catalonia Barcelona Spain College of Computer and Big Data Fuzhou University Fuzhou University Town No. 2 Wulong Jiangbei Avenue Fuzhou China Digital Media Art Key Laboratory of Sichuan Province Sichuan Conservatory of Music No. 2 Wannianchang Street Chenghua District Sichuan Province Chengdu City China Fuzhou Technology Innovation Center of intelligent Manufacturing information System Minjiang University Fuzhou University Town NO. 200 Xiyuangong Road Fuzhou China Fujian Province University No. 599 Quanxiu Road Juyuanzhou Ecological Civilization District Fujian Province Fuzhou City China College of Electronic Engineering Shandong University of Science and Technology No. 579 Qianwangang Road Huangdao District Qingdao China School of Mechanical and Automotive Engineering Fujian University of Technology No. 33 Xuefu South Road University New District Fujian Province Fuzhou City China

When EEG signals are used to assess the level of student engagement in online teaching tasks, they are often interfered by noise. It is a challenge to effectively remove these noises. Currently, deep learning methods have been applied to the field of EEG denoising. However, existing methods still have some problems. The denoised signals still have obvious noise residue, or the original EEG signals are damaged, and the model fitting speed is too slow. Image dehazing, as a typical denoising task in the field of image enhancement, has achieved great success in recent years. Therefore, inspired by advanced models in this field, we introduce CNN into EEG denoising tasks. In this paper, we take GCANet, an excellent image enhancement model, as an example. The dilated convolutions and gate fusion subnetworks included in GCANet enable more efficient EEG signal denoising. The results demonstrate that the proposed model effectively reduces noise while preserving essential features. Furthermore, in comparison to other state-of-the-art models, our proposed model exhibits enhanced robustness and faster convergence, as evidenced by achieving lower loss values after five epochs. Its good performance provides a new development idea for the field of EEG denoising. © 2023, Taiwan Ubiquitous Information CO LTD. All rights reserved.

关键词： Image enhancement

来源：评论

学校读者我要写书评

暂无评论

A Secure Multi-party Computational Adversary Selection Model Based on Time-Varying of Entropy 3rd

A Secure Multi-party Computational Adversary Selection Model...

引用

3rd International Conference on Machine Learning for Cyber Security, ML4CS 2020

作者： Zhang, Xinyu Liu, YuJun Chen, Yuling Wang, Zhan State Key Laboratory of Public Big Date College of Computer Science and Technology Guizhou University Guiyang China School of Computer Science Nankai University Tianjin China Technical Center of Beijing Customs Beijing China

ISBN: (纸本)9783030622220

In cryptography, secure multi-party computation involves multi-party participant entities, each of them has its own secret input and wants to jointly compute a function through some interaction protocol. In this process, cheating by any of the parties will cause errors in the final result. In order to prevent parties from cheating, many experts have proposed protocols with semi-honest security, malicious security and covert security. However, existing protocols can only detect and deal with malicious attackers after being attacked. This passive approach will not only consume more resources, but also cause excessive losses for participants. To solve the above problems, this paper proposes an optimal participant selection model based on the time-varying entropy. Over time, we evaluate the safety entropy of the secure multi-party computation model. Once the entropy value exceeds the expectation, we will reselect the adversary through the reputation mechanism, thus always ensuring the safe state of the model. Our model can better predict the unsafe state of the system, and the security of the model can be guaranteed by the choice of the adversary. At this same time, it has better efficiency in the design of password protocol. © 2020, Springer Nature Switzerland AG.

关键词： Entropy

来源：评论

学校读者我要写书评

暂无评论

Ppdrm: Privacy-Preserving Drm Training and Classification on the Cloud

SSRN

引用

SSRN 2023年

作者： Xu, Lida Tian, Chengliang Zhang, Guoyan Li, Leibo Tian, Weizhong Zhang, Yan College of Computer Science and Technology Qingdao University Qingdao266071 China School of Cyber Science and Technology Shandong University Shandong Ji’nan250101 China Key Laboratory of Cryptologic Technology and Information Security Ministry of Education Shandong Ji’nan250101 China Shandong Institute of Blockchain Shandong Ji’nan250101 China College of Big Data and Internet Shenzhen Technology University Guangdong Shenzhen518118 China College of Electromechanical Engineering Qingdao University of Science and Technology Qingdao266061 China

The prevalence of cloud computing greatly promotes the development of artificial intelligence(AI). Accurately and efficiently classify data on the cloud is a classic AI task. The recent Discriminative Ridge Machine(DRM) is a new type of supervised classifier which outperforms many other classic classifiers in accuracy and efficiency, especially for high-dimensional data or unbalanced ***, the DRM training and classification task is tedious for resource-constrained data owners. Motivated by this, in this paper, we study the outsourcing of DRM training and classification within a cloud environment and propose a quantum secure and efficient solution for this problem. Specifically, based on the two non-colluding servers model, we adopt the quantum-immune CKKS cryptosystem to design a series of homomorphic security protocols for matrix and argmin operations and realize the correct model training and data classification without leaking the privacy of training dataset, the DRM model, the query data and the classification results. We argue the $\mathcal{CPA}$ secure of our design and evaluate the accuracy and efficiency from theory and experiment perspectives. Our theoretical and experimental analysis demonstrates that our design can achieve the same classification accuracy as the origin plaintext-based DRM classifier and the query-response time is reasonable and acceptable. © 2023, The Authors. All rights reserved.

关键词： Classification (of information)

来源：评论

学校读者我要写书评

暂无评论

KA2ER: Knowledge Adaptive Amalgamation of ExpeRts for Medical Images Segmentation

arXiv

引用

arXiv 2024年

作者： Gao, Shangde Fu, Yichao Liu, Ke Xu, Hongxia Wu, Jian College of Computer Science and Technology Zhejiang University Hangzhou China Liangzhu Laboratory Zhejiang University Hangzhou China WeDoctor Holdings Limited Hangzhou China State Key Laboratory of Transvascular Implantation Devices of The Second Affiliated Hospital Zhejiang University School of Medicine Hangzhou China School of Public Health Zhejiang University Hangzhou China

Recently, many foundation models for medical image analysis such as MedSAM, SwinUNETR have been released and proven to be useful in multiple tasks. However, considering the inherent heterogeneity and inhomogeneity of real-world medical data, directly applying these models to specific medical image segmentation tasks often leads to negative domain shift effects, which can severely weaken the model’s segmentation capabilities. To this end, we propose an adaptive amalgamation knowledge framework that aims to train a versatile foundation model to handle the joint goals of multiple expert models, each specialized for a distinct task. Specifically, we first train an nnUNet-based expert model for each task, and reuse the pre-trained SwinUNTER as the target foundation model. Then, the input data for all challenging tasks are encoded in the foundation model and the expert models, respectively, and their backbone features are jointly projected into the adaptive amalgamation layer. Within the hidden layer, the hierarchical attention mechanisms are designed to achieve adaptive merging of the target model to the hidden layer feature knowledge of all experts, which significantly reduces the domain shift arising from the inter-task differences. Finally, the gold amalgamated features and the prompt features are fed into the mask decoder to obtain the segmentation results. Extensive experiments conducted in these challenging tasks demonstrate the effectiveness and adaptability of our foundation model for real-world medical image segmentation. Copyright © 2024, The Authors. All rights reserved.

关键词： Image segmentation

来源：评论

学校读者我要写书评

暂无评论

Impacts of Radiance Quantization on Surface Mapping: Comparisons among the Landsat Sensors 28

Impacts of Radiance Quantization on Surface Mapping: Compari...

引用

28th International Conference on Geoinformatics, Geoinformatics 2021

作者： Chen, Feng Zhang, Ziqun Zhang, Wenhao Song, Yuejun College of Computer and Information Engineering Xiamen University of Technology Xiamen361024 China Big Data Institute of Digital Natural Disaster Monitoring in Fujian Xiamen University of Technology Xiamen361024 China Jiangxi Academy of Water Science and Engineering Jiangxi Key Laboratory of Soil Erosion and Prevention Nanchang330029 China

ISBN: (纸本)9781665420747

Radiance quantization and dynamic ranges are considered the basic characteristics of a sensor in radiometry. However, their impacts on surface mapping have not been fully investigated and understood (e.g., for different Landsat sensors), while much attention has been paid to the impacts related to other aspects (e.g., spatial resolution, the wavelength of spectral channels, and several spectral channels). In this paper, a synthesized data collection of multispectral reflectance and radiance (with and without considering actual quantization) for the Landsat sensors from hyperspectral records was generated. Based on the synthesized data, the impacts of radiance quantization were discussed in terms of channel reflectance and spectral indices. As the preliminary findings show, the quantization effects on channel reflectance vary among channels, presenting different impacts on spectral indices correspondingly, which affect the measures of between-sensor difference. It also suggests that the effects of quantization may relate to land surface conditions (e.g., land use and land cover) in addition to the sensor characteristics. © 2021 IEEE.

关键词： Landsat

来源：评论

学校读者我要写书评

暂无评论

Bitcoin-Based Anti-collusion Fair Payments for Outsourcing Computations in Cloud Computing 1st

Bitcoin-Based Anti-collusion Fair Payments for Outsourcing C...

引用

1st International Conference on Security and Privacy in Digital Economy, SPDE 2020

作者： Zhang, Duo Zhang, Xiaodong Gao, Sheng Tian, Youliang State Key Laboratory of Public Big Data College of Computer Science and Technology Guizhou University GuiYang China School of Mathematics and Statistics Guizhou University GuiYang China College of Mathematics and Computer Science Liupanshui Normal College Liupanshui China School of Information Central University of Finance and Economics Beijing China

ISBN: (纸本)9789811591280

As an attractive cloud computing model, outsourcing computation often has a fair payment issues. In cloud computing, resource-constrained users outsource tasks and pay for them. However, the traditional outsourcing computation model can hardly resist the threat of collusion between calculators. In this paper, we propose a new scheme of Bitcoin-based anti-collusion fair payments for outsourcing computations (BAPay), give the architecture of BAPay, security requirements, and describe the design details. Besides, the scheme guarantees that an honest calculator will be paid for doing the calculations, regardless of the outsourcer’s behaviour. In addition, the security analysis shows that BAPay achieves completeness and fairness. © 2020, Springer Nature Singapore Pte Ltd.

关键词： Cloud computing

来源：评论

学校读者我要写书评

暂无评论

Visible-Thermal Multiple Object Tracking: Large-scale Video dataset and Progressive Fusion Approach

arXiv

引用

arXiv 2024年

作者： Zhu, Yabin Wang, Qianwu Li, Chenglong Tang, Jin Huang, Zhixiang The Key Laboratory of Intelligent Computing and Signal Processing Ministry of Education Anhui University School of Public Safety and Emergency Management Anhui University of Science and Technology Hefei231131 China School of Artificial Intelligence Anhui University Hefei230601 China Information Materials and Intelligent Sensing Laboratory of Anhui Province Anhui Provincial Key Laboratory of Security Artificial Intelligence School of Artificial Intelligence Anhui University Hefei230601 China Information Materials and Intelligent Sensing Laboratory of Anhui Province Anhui Provincial Key Laboratory of Multimodal Cognitive Computation School of Computer Science and Technology Anhui University Hefei230601 China Center for Big Data and Population Health of IHM China

The complementary benefits from visible and thermal infrared data are widely utilized in various computer vision task, such as visual tracking, semantic segmentation and object detection, but rarely explored in Multiple Object Tracking (MOT). In this work, we contribute a large-scale Visible-Thermal video benchmark for MOT, called VT-MOT. VT-MOT has the following main advantages. 1) The data is large scale and high diversity. VT-MOT includes 582 video sequence pairs, 401k frame pairs from surveillance, drone, and handheld platforms. 2) The cross-modal alignment is highly accurate. We invite several professionals to perform both spatial and temporal alignment frame by frame. 3) The annotation is dense and high-quality. VT-MOT has 3.99 million annotation boxes annotated and double-checked by professionals, including heavy occlusion and object re-acquisition (object disappear and reappear) challenges. To provide a strong baseline, we design a simple yet effective tracking framework, which effectively fuses temporal information and complementary information of two modalities in a progressive manner, for robust visible-thermal MOT. A comprehensive experiment are conducted on VT-MOT and the results prove the superiority and effectiveness of the proposed method compared with state-of-the-art methods. From the evaluation results and analysis, we specify several potential future directions for visible-thermal MOT. The project is released in https://***/wqw123wqw/PFTrack. Copyright © 2024, 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

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：