检索结果-内蒙古大学图书馆

12th International Conference on Image and Graphics, ICIG 2023

作者： Xiao, Yongbiao Li, Hui Cheng, Chunyang Song, Xiaoning International Joint Laboratory on Artificial Intelligence of Jiangsu Province School of Artificial Intelligence and Computer Science Jiangnan University Wuxi214122 China

ISBN: (纸本)9783031463044

Infrared and visible image fusion task aims to generate a fused image which contains salient features and rich texture details from multi-source images. However, under complex illumination conditions, few algorithms pay attention to the edge information of local regions which is crucial for downstream tasks. To this end, we propose a fusion network based on the local edge enhancement, named LE2Fusion. Specifically, a local edge enhancement (LE2) module is proposed to improve the edge information under complex illumination conditions and preserve the essential features of image. For feature extraction, a multi-scale residual attention (MRA) module is applied to extract rich features. Then, with LE2, a set of enhancement weights are generated which are utilized in feature fusion strategy and used to guide the image reconstruction. To better preserve the local detail information and structure information, the pixel intensity loss function based on the local region is also presented. The experiments demonstrate that the proposed method exhibits better fusion performance than the state-of-the-art fusion methods on public datasets. © The Author(s), under exclusive license to Springer Nature Switzerland AG 2023.

关键词： Image fusion

来源：评论

学校读者我要写书评

暂无评论

Semi-supervised Counting of Grape Berries in the Field Based on Density Mutual Exclusion

引用

Plant Phenomics 2023年第4期5卷 825-835页

作者： Yanan Li Yuling Tang Yifei Liu Dingrun Zheng School of Computer Science and Engineering School of Artificial IntelligenceWuhan Institute of TechnologyWuhan 430205China Hubei Key Laboratory of Intelligent Robot Wuhan Institute of TechnologyWuhan 430073China

Automated counting of grape berries has become one of the most important tasks in grape yield ***,dense distribution of berries and the severe occlusion between berries bring great challenges to counting algorithm based on deep *** collection of data required for model training is also a tedious and expensive *** address these issues and cost-effectively count grape berries,a semi-supervised counting of grape berries in the field based on density mutual exclusion(CDMENet)is *** algorithm uses VGG16 as the backbone to extract image *** tasks based on density mutual exclusion are *** tasks exploit the spatial distribution pattern of grape berries in density levels to make full use of unlabeled *** addition,a density difference loss is *** feature representation is enhanced by amplifying the difference of features between different density *** experimental results on the field grape berry dataset show that CDMENet achieves less counting *** with the state of the arts,coefficient of determination(R^(2))is improved by 6.10%,and mean absolute error and root mean square error are reduced by 49.36%and 54.08%,*** code is available at.

关键词： Grape mutual backbone

来源：评论

学校读者我要写书评

暂无评论

Security and Privacy in Solar Insecticidal Lamps Internet of Things:Requirements and Challenges

引用

IEEE/CAA Journal of Automatica Sinica 2024年第1期11卷 58-73页

作者： Qingsong Zhao Lei Shu Kailiang Li Mohamed Amine Ferrag Ximeng Liu Yanbin Li the College of Artificial Intelligence Nanjing Agricultural UniversityNanjing 210031 the State Key Laboratory for Novel Software Technology Nanjing UniversityNanjing 210023 the State Key Laboratory of Cryptology Beijing 100878China IEEE the School of Engineering University of LincolnLincoln LN67TSUK the Artificial Intelligence and Digital Science Research Center Technology Innovation InstituteMasdar City 9639UAE the College of Computer and Data Science Fuzhou UniversityFuzhou 350108China

Solar insecticidal lamps(SIL) can effectively control pests and reduce the use of pesticides. Combining SIL and Internet of Things(IoT) has formed a new type of agricultural IoT,known as SIL-IoT, which can improve the effectiveness of migratory phototropic pest control. However, since the SIL is connected to the Internet, it is vulnerable to various security *** issues can lead to serious consequences, such as tampering with the parameters of SIL, illegally starting and stopping SIL,etc. In this paper, we describe the overall security requirements of SIL-IoT and present an extensive survey of security and privacy solutions for SIL-IoT. We investigate the background and logical architecture of SIL-IoT, discuss SIL-IoT security scenarios, and analyze potential attacks. Starting from the security requirements of SIL-IoT we divide them into six categories, namely privacy, authentication, confidentiality, access control, availability,and integrity. Next, we describe the SIL-IoT privacy and security solutions, as well as the blockchain-based solutions. Based on the current survey, we finally discuss the challenges and future research directions of SIL-IoT.

关键词： Challenges Internet of Things(IoT) privacy and security security requirements solar insecticidal lamps(SIL)

来源：评论

学校读者我要写书评

暂无评论

Selective Depth Attention Networks for Adaptive Multiscale Feature Representation

IEEE Transactions on Artificial Intelligence

引用

IEEE Transactions on artificial intelligence 2024年第10期5卷 5064-5074页

作者： Guo, Qingbei Wu, Xiao-Jun Xu, Tianyang Si, Tongzhen Hu, Cong Tian, Jinglan University of Jinan Shandong Provincial Key Laboratory of Network-based Intelligent Computing Jinan250022 China Jiangnan University School of Artificial Intelligence and Computer Science Wuxi214122 China

Existing multiscale methods lead to a risk of just increasing the receptive field sizes while neglecting small receptive fields. Thus, it is a challenging problem to effectively construct adaptive neural networks for recognizing various spatial-scale objects. To tackle this issue, we first introduce a new attention dimension, i.e., depth, in addition to existing attentions such as channel-attention, spatial-attention, branch-attention, and self-attention. We present a novel selective depth attention network to treat multiscale objects symmetrically in various vision tasks. Specifically, the blocks within each stage of neural networks, including convolutional neural networks (CNNs), e.g., ResNet, SENet, and Res2Net, and vision transformers (ViTs), e.g., PVTv2, output the hierarchical feature maps with the same resolution but different receptive field sizes. Based on this structural property, we design a depthwise building module, namely an selective depth attention (SDA) module, including a trunk branch and a SE-like attention branch. The block outputs of the trunk branch are fused to guide their depth attention allocation through the attention branch globally. According to the proposed attention mechanism, we dynamically select different depth features, which contributes to adaptively adjusting the receptive field sizes for the variable-sized input objects. Moreover, our method is orthogonal to multiscale networks and attention networks, so-called SDA-xNet. Extensive experiments demonstrate that the proposed SDA method significantly improves the original performance as a lightweight and efficient plug-in on numerous computer vision tasks, e.g., image classification, object detection, and instance segmentation. © 2020 IEEE.

关键词： Deep neural networks

来源：评论

学校读者我要写书评

暂无评论

Heterogeneous Image Knowledge Driven Visual Perception

引用

IEEE/CAA Journal of Automatica Sinica 2024年第1期11卷 255-257页

作者： Lan Yan Wenbo Zheng Fei-Yue Wang IEEE the College of Computer Science and Electronic Engineering Hunan UniversityChangsha 410082China the School of Computer Science and Artificial Intelligence Wuhan University of TechnologyWuhan 430070China the State Key Laboratory for Management and Control of Complex Systems Institute of AutomationChinese Academy of SciencesBeijing 100190China

Dear Editor,This letter is concerned with visual perception closely related to heterogeneous *** the huge challenge brought by different image modalities,we propose a visual perception framework based on heterogeneous image knowledge,i.e.,the domain knowledge associated with specific vision tasks,to better address the corresponding visual perception problems.

关键词： Visual visual knowledge

来源：评论

学校读者我要写书评

暂无评论

On the Emergence of Cross-Task Linearity in Pretraining-Finetuning Paradigm 41

On the Emergence of Cross-Task Linearity in Pretraining-Fine...

引用

41st International Conference on Machine Learning, ICML 2024

作者： Zhou, Zhanpeng Chen, Zijun Chen, Yilan Zhang, Bo Yan, Junchi School of Artificial Intelligence Department of Computer Science and Engineering MoE Lab of AI Shanghai Jiao Tong University Shanghai China Shanghai Artificial Intelligence Laboratory China Computer Science and Engineering University of California San Diego United States

The pretraining-finetuning paradigm has become the prevailing trend in modern deep *** this work, we discover an intriguing linear phenomenon in models that are initialized from a common pretrained checkpoint and finetuned on different tasks1, termed as Cross-Task Linearity (CTL).Specifically, we show that if we linearly interpolate the weights of two finetuned models, the features in the weight-interpolated model are often approximately equal to the linear interpolation of features in two finetuned models at each *** provide comprehensive empirical evidence supporting that CTL consistently occurs for finetuned models that start from the same pretrained *** conjecture that in the pretraining-finetuning paradigm, neural networks approximately function as linear maps, mapping from the parameter space to the feature *** on this viewpoint, our study unveils novel insights into explaining model merging/editing, particularly by translating operations from the parameter space to the feature ***, we delve deeper into the root cause for the emergence of CTL, highlighting the role of *** released our source code at https://***/zzp1012/Cross-Task-Linearity. Copyright 2024 by the author(s)

关键词： Deep learning

来源：评论

学校读者我要写书评

暂无评论

A unified user behavior model for trajectory-based tasks with different types of path constraints

引用

science China(Information sciences) 2023年第10期66卷 316-318页

作者： Hao ZHANG Jin HUANG Huawei TU Feng TIAN Guozhong DAI Hongan WANG Beijing Key Laboratory of Human-Computer Interaction Institute of Software Chinese Academy of Sciences School of Computer Science and Technology University of Chinese Academy of Sciences School of Artificial Intelligence University of Chinese Academy of Sciences Department of Computer Science and Information Technology La Trobe University

An important research branch of human-computer interaction(HCI) is to develop predictive models for human performance in fundamental interactions [1]. On today's graphical user interface(GUI), users often implicitly perform various trajectory-based interactions, such as navigating through menus [2], entering the boundary of a button,

关键词：

来源：评论

学校读者我要写书评

暂无评论

Generating Editable Head Avatars with 3D Gaussian GANs

Generating Editable Head Avatars with 3D Gaussian GANs

引用

2025 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2025

作者： Li, Guohao Yang, Hongyu Men, Yifang Huang, Di Li, Weixin Yang, Ruijie Wang, Yunhong School of Computer Science and Engineering Beihang University Beijing China School of Artificial Intelligence Beihang University Beijing China Shanghai Artificial Intelligence Laboratory Shanghai China

ISBN: (纸本)9798350368741

Generating animatable and editable 3D head avatars is essential for various applications in computer vision and graphics. Traditional 3D-aware generative adversarial networks (GANs), often using implicit fields like Neural Radiance Fields (NeRF), achieve photorealistic and view-consistent 3D head synthesis. However, these methods face limitations in deformation flexibility and editability, hindering the creation of lifelike and easily modifiable 3D heads. We propose a novel approach that enhances the editability and animation control of 3D head avatars by incorporating 3D Gaussian Splatting (3DGS) as an explicit 3D representation. This method enables easier illumination control and improved editability. Central to our approach is the Editable Gaussian Head (EG-Head) model, which combines a 3D Morphable Model (3DMM) with texture maps, allowing precise expression control and flexible texture editing for accurate animation while preserving identity. To capture complex non-facial geometries like hair, we use an auxiliary set of 3DGS and tri-plane features. Extensive experiments demonstrate that our approach delivers high-quality 3D-aware synthesis with state-of-the-art controllability. Our code and models are available at https://***/liguohao96/EGG3D. © 2025 IEEE.

关键词： 3D-aware GAN Image video synthesis

来源：评论

学校读者我要写书评

暂无评论

Density estimation-based method to determine sample size for random sample partition of big data

引用

Frontiers of computer science 2024年第5期18卷 57-70页

作者： Yulin HE Jiaqi CHEN Jiaxing SHEN Philippe FOURNIER-VIGER Joshua Zhexue HUANG Guangdong Laboratory of Artificial Intelligence and Digital Economy(SZ) Shenzhen 518107China College of Computer Science and Software Engineering Shenzhen UniversityShenzhen 518060China Department of Computing and Decision Sciences Lingnan UniversityHong Kong 999077China

Random sample partition(RSP)is a newly developed big data representation and management model to deal with big data approximate computation *** research and practical applications have confirmed that RSP is an efficient solution for big data processing and ***,a challenge for implementing RSP is determining an appropriate sample size for RSP data *** a large sample size increases the burden of big data computation,a small size will lead to insufficient distribution information for RSP data *** address this problem,this paper presents a novel density estimation-based method(DEM)to determine the optimal sample size for RSP data ***,a theoretical sample size is calculated based on the multivariate Dvoretzky-Kiefer-Wolfowitz(DKW)inequality by using the fixed-point iteration(FPI)***,a practical sample size is determined by minimizing the validation error of a kernel density estimator(KDE)constructed on RSP data blocks for an increasing sample ***,a series of persuasive experiments are conducted to validate the feasibility,rationality,and effectiveness of *** results show that(1)the iteration function of the FPI method is convergent for calculating the theoretical sample size from the multivariate DKW inequality;(2)the KDE constructed on RSP data blocks with sample size determined by DEM can yield a good approximation of the probability density function(p.d.f);and(3)DEM provides more accurate sample sizes than the existing sample size determination methods from the perspective of *** demonstrates that DEM is a viable approach to deal with the sample size determination problem for big data RSP implementation.

关键词： random sample partition big data sample size Dvoretzky-Kiefer-Wolfowitz inequality kerneldensity estimator probability density function

来源：评论

学校读者我要写书评

暂无评论

Label distribution similarity-based noise correction for crowdsourcing

引用

Frontiers of computer science 2024年第5期18卷 71-82页

作者： Lijuan REN Liangxiao JIANG Wenjun ZHANG Chaoqun LI School of Computer Science China University of GeosciencesWuhan 430074China Key Laboratory of Artificial Intelligence Ministry of EducationShanghai 200240China School of Mathematics and Physics China University of GeosciencesWuhan 430074China

In crowdsourcing scenarios,we can obtain each instance's multiple noisy labels from different crowd workers and then infer its integrated label via label *** spite of the effectiveness of label aggregation methods,there still remains a certain level of noise in the integrated ***,some noise correction methods have been proposed to reduce the impact of noise in recent ***,to the best of our knowledge,existing methods rarely consider an instance's information from both its features and multiple noisy labels simultaneously when identifying a noise *** this study,we argue that the more distinguishable an instance's features but the noisier its multiple noisy labels,the more likely it is a noise *** on this premise,we propose a label distribution similarity-based noisecorrection(LDSNC)*** measure whether an instance's features are distinguishable,we obtain each instance's predicted label distribution by building multiple classifiers using instances'features and their integrated *** measure whether an instance's multiple noisy labels are noisy,we obtain each instance's multiple noisy label distribution using its multiple noisy ***,we use the Kullback-Leibler(KL)divergence to calculate the similarity between the predicted label distribution and multiple noisy label distribution and define the instance with the lower similarity as a noise *** extensive experimental results on 34 simulated and four real-world crowdsourced datasets validate the effectiveness of our method.

关键词： crowdsourcing learning noise correction label distribution similarity kullback-leibler divergence

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：