检索结果-内蒙古大学图书馆

22nd IEEE International Symposium on Parallel and Distributed Processing with Applications, ISPA 2024

作者： Sun, Guoqing Lian, Zhichao Nanjing University of Science and Technology School of Computer Science and Engineering Nanjing China Nanjing University of Science and Technology School of Cyber Science and Engineering Nanjing China

ISBN: (纸本)9798331509712

With the rapid development of Deepfake synthesis technology in recent years, our cybersecurity and individual privacy face challenges. In pursuit of robust Deepfake detection, researchers have tried to use temporal cues in videos, employing models such as RNNs and 3D convolutional networks. Despite these efforts, there remains ample opportunity for improvement within these models. In this paper, we introduced an approach utilizing the Video Vision Transformer, which is based on the decomposition of Spatial-Temporal attention mechanisms, for the detection of video face forgery. This method is designed to capture spatial artifacts and temporal inconsistencies. Besides, difference module is introduced to screen features and reduce the interference of natural factors on the model. To enable robust Deepfake detection. We have conducted extensive experiments across some datasets, including FaceForensics++, DFDC and WildDeepfake datasets. Which demonstrates the validity of the model. © 2024 IEEE.

关键词： Deepfake detection Difference module Spatial-Temporal Attention Mechanism

来源：评论

学校读者我要写书评

暂无评论

CMCF-Net: An End-to-End Context Multiscale Cross-Fusion Network for Robust Copy-Move Forgery Detection

引用

IEEE Transactions on Multimedia 2024年 26卷 6090-6101页

作者： Xiong, Lizhi Xu, Jianhua Yang, Ching-Nung Zhang, Xinpeng Nanjing University of Information Science and Technology School of Computer Science Nanjing210044 China Jiangsu Collaborative Innovation Center on Atmospheric Environment and Equipment Technology Engineering Research Center of Digital Forensics Ministry of Education Nanjing210044 China Nanjing University of Information Science and Technology Software Engineering Nanjing210044 China National Dong Hwa University Department of Computer Science and Information Engineering Hualien974301 Taiwan Fudan University School of Computer Science Shanghai200433 China

Image copy-move forgery detection (CMFD) has become a challenging problem due to increasingly powerful editing software that makes forged images increasingly realistic. Existing algorithms that directly connect multiple scales of features in the encoder part may not effectively aggregate contextual information, resulting in poor performance. In this paper, an end-to-end context multiscale cross-fusion network (CMCF-Net) is proposed to detect image copy-move forgery. The proposed network consists of a multiscale feature extraction fusion (MSF) module and a multi-information fusion decoding (MFD) module. Multiscale information is efficiently extracted and fused in the MSF module utilizing stacked-scale feature fusion, which improves the network's forgery localization ability on objects of different scales. The MFD module employs contextual information combination and weighted fusion of multiscale information to guide the network in obtaining relevant clues from correlated information at multiple different scales. Experimental results and analysis have demonstrated that the proposed CMCF-Net achieves the best localization results with higher robustness. © 1999-2012 IEEE.

关键词： Feature extraction

来源：评论

学校读者我要写书评

暂无评论

Double-Click: dual-click iteration and dual-branch networks for interactive segmentation 16

Double-Click: dual-click iteration and dual-branch networks ...

引用

2024 16th International Conference on Graphics and Image Processing, ICGIP 2024

作者： Jiang, Mingcheng Wang, Di Ren, Benzhe School of Computer Science and Engineering Nanjing University of Science and Technology Nanjing210094 China

ISBN: (数字)9781510688780

ISBN: (纸本)9781510688773

The main aim of Interactive image segmentation is to accurately segment regions of interest by utilizing simple user-provided interactive information. Currently, the most popular training approach involves using the previous segmentation result as supplementary information for the next segmentation process to establish a connection between the two interactions. This iterative training method transfers all the information carried by interactions to the global level, without considering its impact at the local level. This will affect the final segmentation results and the training efficiency. To address this problem, we propose an iterative training strategy called Double-Click to involve cropping the image based on the newly added clicks and then separately feeding the cropped image and the newly added clicks into a smaller branch network to obtain predictions for the local region. Subsequently, the locally segmented result is fused with the previous segmentation result to achieve the final outcome. Then, the refined segmentation results are used to guide the next segmentation, ensuring that more precise results guide subsequent segmentations while maintaining the guidance information of the clicks on the global image. Our training strategy improves the training effectiveness by handling two clicks in an iterative process. The accuracy of the segmentation is ensured through both local and global predictions. Additionally, we designed a dual-branch network to accommodate our proposed training strategy. The effectiveness of this strategy has been tested on four validated on four benchmarks, namely GrabCut, Berkeley, SBD, and DAVIS. Experimental results demonstrate that the Double-Click method achieves better performance with fewer interactions compared to previous methods. © 2025 SPIE.

关键词： Semantic Segmentation

来源：评论

学校读者我要写书评

暂无评论

A novel large kernel two-branch semantic segmentation network 24

A novel large kernel two-branch semantic segmentation networ...

引用

4th International Conference on Signal Processing and Communication technology, SPCT 2024

作者： Zhang, Xinrui Fu, Peng School of Computer Science and Engineering Nanjing University of Science and Technology Jiangsu Nanjing China

ISBN: (纸本)9798400710636

For semantic branching in a two-branch network structure, it is crucial to quickly improve the feeling field, in addition, the feature fusion interaction of two-branching needs to take into account the structural and multi-scale information fusion of features under different feeling fields. In this paper, in order to cope with this challenge, we propose to accumulate the receptive field in dense convolution using separable large kernel convolution through an adaptive semantic extraction module, and design a selectable large kernel attention module for the problem of mismatch of receptive field of feature information. Finally, this paper designs a multiscale feature extraction module that fuses local and global information, and uses large kernel dense separable convolution in the module with the ability to extract high-dimensional semantics serially and capture multiscale in parallel. © 2024 Copyright held by the owner/author(s).

关键词： Semantic Segmentation

来源：评论

学校读者我要写书评

暂无评论

A Survey of the Application of Neural Networks to Event Extraction

引用

Tsinghua science and technology 2025年第2期30卷 748-768页

作者： Jianye Xie Yulan Zhang Huaizhen Kou Xiaoran Zhao Zhikang Feng Lekang Song Weiyi Zhong College of Computer Science and Technology China University of Petroleum(East China)Qingdao 266000China Shandong Provincial University Laboratory for Protected Horticulture Weifang University of Science and TechnologyWeifang 261000China School of Computer Science and Engineering Nanjing University of Science and TechnologyNanjing 210000China School of Computer Science Qufu Normal UniversityRizhao 276800China

Event extraction is an important part of natural language information extraction,and it’s widely employed in other natural language processing tasks including question answering and machine reading ***,there is a lack of recent comprehensive survey papers on event *** the past few years,numerous high-quality and innovative event extraction methods have been proposed,making it necessary to consolidate these new developments with previous work in order to provide a clear overview for researchers and serve as a reference for future *** addition,event detection is a fundamental sub-task in event extraction,previous survey papers have often overlooked the related work on event ***,this paper aims to bridge these gaps by presenting a comprehensive survey of event extraction,including recent advancements and an analysis of previous research on event *** resources for event extraction are first introduced in this research,and then the numerous neural network models currently employed in event extraction tasks are divided into four types:word sequence-based methods,graph-based neural network methods,external knowledge-based approaches,and prompt-based *** compare and contrast them in depth,pointing out the flaws and difficulties with existing ***,we discuss the future of event extraction development.

关键词： event extraction natural language processing event extraction methods graph neural network prompt-based learning

来源：评论

学校读者我要写书评

暂无评论

Recommendation Algorithm Based on Social Influence Diffusion 17

Recommendation Algorithm Based on Social Influence Diffusion

引用

IEEE Congress on Cybermatics: 17th IEEE International Conference on Internet of Things, iThings 2024, 20th IEEE International Conference on Green Computing and Communications, GreenCom 2024, 17th IEEE International Conference on Cyber, Physical and Social Computing, CPSCom 2024, 10th IEEE International Conference on Smart Data, SmartData 2024

作者： Jin, Xin Chen, Shiyu Nanjing University of Science and Technology School of Mechanical Engineering Nanjing China Nanjing University of Science and Technology School of Computer Science and Engineering Nanjing China

ISBN: (纸本)9798350351637

Social influence is everywhere in people's lives. It controls users' activities in social networks. Understanding the mechanism of interactions between users in social networks can benefit various applications, such as recommendations. Traditional recommendation systems based on collaborative filtering suppose that the behaviors of users are independent of each other and ignore the influence relationships among users. However, different historical behaviors and social relationships of users result in different influences among users. First, the complete user behavior sequences are fed into a recurrent neural network to capture long term patterns of user interest in items. Secondly, a self-attention network is used to obtain short lived interests of users. Then the outputs of the recurrent neural network as queries are fed into another self-attention network to obtain user representations that combine transient effect of and long term patterns of users' interests, thereby predicting the ratings of users to items. Finally, the proposed recommendation system is used to recommend products that need to be promoted to the maximum number of users. The proposed method is verified to outperform the state-of-the-art algorithms on three real-world datasets. © 2024 IEEE.

关键词： Chatbots

来源：评论

学校读者我要写书评

暂无评论

Deep Contrastive Clustering for Protocol Analysis in 5G Network engineering 4

Deep Contrastive Clustering for Protocol Analysis in 5G Netw...

引用

4th International Conference on Communication technology and Information technology, ICCTIT 2024

作者： Lu, Jianhong Zhang, Kun Nanjing University of Science and Technology School of Cyber Science and Engineering Nanjing China Nanjing University of Science and Technology School of Computer and Engineering Nanjing China

ISBN: (纸本)9798331528973

With the widespread adoption of 5G networks in satellite communication and vehicular communication systems, the complexity of communication and signaling interactions has significantly increased. Traditional protocol analysis methods primarily focus on high-level structured data, often neglecting the low-level binary data critical for understanding 5G-specific protocols. This paper proposes a novel approach that employs a deep contrastive clustering framework to analyze and classify unknown protocols in 5G communication networks. The proposed method leverages depthwise separable convolutions for feature extraction, transforming binary protocol data into structured feature representations while significantly reducing computational overhead. Additionally, by integrating contrastive learning into the clustering process, the framework enhances the discrimination of protocol embeddings through a positive-negative sample feedback mechanism, ensuring more accurate clustering results. Additionally, this paper collected and analyzed a comprehensive dataset of 5 G protocols, which includes wireless access layer protocols and core network communication protocols. © 2024 IEEE.

关键词： Satellite communication systems

来源：评论

学校读者我要写书评

暂无评论

BiMoeFormer: A Novel Transformer-Based Framework for 3D Human Motion Prediction with Bi-Level Routing Attention and Mixture of Experts 5

BiMoeFormer: A Novel Transformer-Based Framework for 3D Huma...

引用

5th International Seminar on Artificial Intelligence, Networking and Information technology, AINIT 2024

作者： Zhang, Wei Nanjing University of Science and Technology School of Computer Science and Engineering Nanjing China

ISBN: (纸本)9798350385557

We introduce a novel BiMoeFormer based on the Transformer architecture for 3D human motion prediction. Previous approaches primarily focus on the relationships between body joints in human poses, while neglecting their independencies. Our method addresses this by incorporating a Bi-Level Routing Attention (BRA) mechanism, which emphasizes balancing the correlation and independence among these joints. More specifically, BRA selects the top k most relevant joints for each joint, enhancing the overall modeling capability. Moreover, to accurately capture the diversity and complexity of human motion, we employ a Mixture of Experts (MoE) architecture to enhance the expressiveness of the Transformer. This allows us to control the model complexity by sharing the MoE across blocks. Comprehensive evaluations on the Human3.6M, AMASS, and 3DPW datasets demonstrate that our method consistently outperforms state-of-the-art baselines. © 2024 IEEE.

关键词： Motion estimation

来源：评论

学校读者我要写书评

暂无评论

An Enhanced Vision-Language Pre-Training Approach for Scene Text Detection 5

An Enhanced Vision-Language Pre-Training Approach for Scene ...

引用

5th International Symposium on computer engineering and Intelligent Communications, ISCEIC 2024

作者： Gu, Tao Zhang, Chongyang School of Computer Science and Engineering Nanjing University of Science and Technology Nanjing China

ISBN: (纸本)9798331518677

Recent advancements in Vision-Language Pre-training (VLP) techniques have greatly improved performance in Scene Text Detection tasks by leveraging the rich visual and textual content in scene text images. We propose an innovative integration of contrastive learning with Masked Language Modeling (MLM) and Masked Image Modeling (MIM), inspired by the Masked Autoencoder (MAE) approach. This synthesis enhances self-supervised learning by combining discriminative and generative learning. Notably, we introduce masked image modeling for text detection, enabling effective representation learning in text image regions. To distinguish our approach from traditional BERT's MLM, we develop a custom tokenizer tailored for scene text detection. Our pre-trained model aligns visual and textual information, resulting in improved performance of existing scene text detectors. Extensive experiments on datasets such as ICDAR2015 show that our method significantly outperforms previous pre-training approaches. © 2024 IEEE.

关键词： Contrastive Learning

来源：评论

学校读者我要写书评

暂无评论

Forecasting Cloud Workload through Reinforcement Learning-Based Ensemble Model 4

Forecasting Cloud Workload through Reinforcement Learning-Ba...

引用

4th International Conference on Electronic Information engineering and computer Communication, EIECC 2024

作者： Kang, Yuliang Li, Yongke Zhang, Gongxuan Nanjing University of Science and Technology School of Computer Science and Engineering Nanjing China

ISBN: (纸本)9798331534622

The escalating demand for cloud computing has intensified the need for accurate workload forecasting to optimize resource allocation and maintain service quality. This paper presents a novel approach employing a reinforcement learning-based ensemble model to enhance the accuracy of workload predictions in cloud environments. The study addresses the challenges posed by complex and dynamic workload patterns, leveraging deep learning techniques to improve prediction outcomes. The proposed model dynamically selects the optimal subset of predictors, demonstrating superior performance on the Google Cluster dataset. The research aims to contribute to more efficient and intelligent resource management in cloud computing. © 2024 IEEE.

关键词： Prediction models

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：