检索结果-内蒙古大学图书馆

arXiv 2019年

作者： Wang, Shaoying Lai, Haijiang Yang, Yifan Yin, Jian School of Electronics and Information Engineering Sun Yat-Sen University Guangzhou China Guangdong Key Laboratory of Big Data Analysis and Processing Guangzhou China School of Data and Computer Science Sun Yat-Sen University Guangzhou China

Deep-networks-based hashing has become a leading approach for large-scale image retrieval, which learns a similarity-preserving network to map similar images to nearby hash codes. The pairwise and triplet losses are two widely used similarity preserving manners for deep hashing. These manners ignore the fact that hashing is a prediction task on the list of binary codes. However, learning deep hashing with listwise supervision is challenging in 1) how to obtain the rank list of whole training set when the batch size of the deep network is always small and 2) how to utilize the listwise supervision. In this paper, we present a novel deep policy hashing architecture with two systems are learned in parallel: A query network and a shared and slowly changing database network. The following three steps are repeated until convergence: 1) the database network encodes all training samples into binary codes to obtain whole rank list, 2) the query network is trained based on policy learning to maximize a reward that indicates the performance of the whole ranking list of binary codes, e.g., mean average precision (MAP), and 3) the database network is updated as the query network. Extensive evaluations on several benchmark datasets show that the proposed method brings substantial improvements over state-of-the-art hashing methods. Copyright © 2019, The Authors. All rights reserved.

关键词： Binary codes

来源：评论

学校读者我要写书评

暂无评论

Deep Learning for Password Guessing and Password Strength Evaluation, A Survey

Deep Learning for Password Guessing and Password Strength Ev...

引用

IEEE International Conference on Trust, Security and Privacy in Computing and Communications (TrustCom)

作者： Tao Zhang Zelei Cheng Yi Qin Qiang Li Lin Shi Guangxi Key Laboratory of Cryptography and Information Security Guilin University of Electronic Technology Guilin China Guangxi Key Laboratory of Trusted Software Guilin University of Electronic Technology Guilin China Guangxi Key Laboratory of Hybrid Computation and IC Design Analysis Guangxi University for Nationalities Nanning China Department of Computer and Information Technology Purdue University West Lafayette USA PLA Information Engineering University Zhengzhou China Henan Key Laboratory of Network Cryptography Technology Zhengzhou China Cloud Computing and Big Data Research Institute China Academy of Information and Communications Technology Beijing China

ISBN: (数字)9781665403924

ISBN: (纸本)9781665403931

Text passwords are the most widely used authentication methods and will also be used in the future. Text passwords can be regarded as meaningful strings, and deep learning methods have an advantage of text processing. LSTM, RNN, GAN and other deep learning models have been using in password guessing and password strength measurements. In the paper, we make a survey on state-of-the-art deep learning methods for password guessing and password strength evaluation, including password pattern extraction, candidate password generation and password strength measurement. Compared with traditional methods, neural networks based methods can achieve better results and performance.

关键词： Deep learning Training Transfer learning Security Gallium nitride Password Text processing

来源：评论

学校读者我要写书评

暂无评论

Early Diagnosis of Alzheimer’s Disease Based on Deep Learning and GWAS 1st

Early Diagnosis of Alzheimer’s Disease Based on Deep Learni...

引用

1st International Workshop on Human Brain and Artificial Intelligence, HBAI 2019, held in conjunction with the 28th International Joint Conference on Artificial Intelligence, IJCAI 2019

作者： Pan, Dan Huang, Yin Zeng, An Jia, Longfei Song, Xiaowei Guangdong Construction Polytechnic Guangzhou510440 China Guangzhou Dazhi Networks Technology Co. Ltd. Guangzhou510000 China Guangdong University of Technology Guangzhou510006 China Guangdong Key Laboratory of Big Data Analysis and Processing Guangzhou510006 China ImageTech Lab Simon Fraser University VancouverV6B 5K3 Canada

ISBN: (纸本)9789811513978

Alzheimer’s disease (AD) is a typical irreversible neurodegenerative disease. At present, the pathogenesis of AD remains elusive and the effective treatment of AD is still a challenge for clinicians. Therefore, early diagnosis is of great importance for the development of new drugs to prevent the progression of AD. With the rapid advancement of neuroimaging technology and deep learning, more and more researchers have turned to deep learning to analyze the brain images for early diagnosis of AD. Plus, studies have demonstrated that it is very likely that the genetic makeup of an individual may influence his/her susceptibility to AD traits. Researchers have begun to identify the genetic biomarkers associated to AD and evaluate the effects of genes upon the changes in the structure and function of the brain of AD patients. In this study, an ensemble model of multi-slice classifiers based on convolutional neural network (CNN) was proposed to make an early diagnosis of AD and at the same time to identify the significant brain regions related to AD. The morphological data of these identified brain regions and the genotype were utilized to carry out genome-wide association studies (GWAS) to explore the potential genetic biomarkers of AD. © 2019, Springer Nature Singapore Pte Ltd.

关键词： Magnetic resonance imaging

来源：评论

学校读者我要写书评

暂无评论

Online Multiple Object Tracking with Reid Feature Extraction Network and Similarity Matrix Network

引用

Journal of Physics: Conference Series 2020年第1期1544卷

作者： Qingge Ji Haoqiang Yu School of Data and Computer Science Sun Yat-sen University Guangzhou 510006 China Guangdong Province Key Laboratory of Big Data Analysis and Processing Guangzhou 510006 China

In multiple object tracking(MOT), data association is a crucial part. By constructing similarity loss matrix for trajectories and detections, they can be matched correspondingly by using Hungarian matching algorithm. However, the similarity loss is often obtained by calculating the Euclidean distance or other handcraft distance metrics of features extracted between objects, which may not be robust enough, resulting in matching inaccuracy. In this paper, we propose a novel MOT method with applying deep learning to feature extraction and data association. We firstly design the appearance feature extraction network(AFN) to learn effective features by training it on a large-scale person re-identification dataset(reid). Then, we propose the similarity matrix estimation network (SMN) to obtain reliable similarity by training it on the public MOT Challenge dataset, MOT17. Additionally, the similarity matrix output of SMN includes the dummy objects, which are used to deal with the association problems of object missing and object appearance between frames. In the end, our proposed MOT method is evaluated on MOT15, MOT17 and ablation study is carried out.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Application of big data Computing Model Based on Semantic data analysis Technology

引用

IOP Conference Series: Materials Science and Engineering 2020年第1期750卷

作者： Ganglong Fan Jian Shen Henan key Laboratory for Big Data Processing & Analytics of Electronic Commerce Luoyang Normal University Luoyang China Higher Education Admissions Office of HENAN Province Zhengzhou China

big data computing model refers to various high-level abstract models extracted and established from the diverse big data computing problems and requirements according to big data's different data characteristics and computational characteristics. In order to improve the data quality, a method of adding semantics to the data is proposed to help users identify big data's pattern. In this way, we can make use of the semantic value of the data, detect big data, and propose a big data computing model through the semantic analysis of the data. The paper presents research on big data computing model based on semantic data analysis technology.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Towards Multi-Pose Guided Virtual Try-On Network

Towards Multi-Pose Guided Virtual Try-On Network

引用

International Conference on Computer Vision (ICCV)

作者： Haoye Dong Xiaodan Liang Xiaohui Shen Bochao Wang Hanjiang Lai Jia Zhu Zhiting Hu Jian Yin School of Data and Computer Science Sun Yat-sen University Guangdong Key Laboratory of Big Data Analysis and Processing Guangzhou P.R.China School of Intelligent Systems Engineering Sun Yat-sen University ByteDance AI Lab. School of Computer Science South China Normal University Guangzhou Key Laboratory of Big Data and Intelligent Education Carnegie Mellon University

ISBN: (数字)9781728148038

ISBN: (纸本)9781728148045

Virtual try-on systems under arbitrary human poses have significant application potential, yet also raise extensive challenges, such as self-occlusions, heavy misalignment among different poses, and complex clothes textures. Existing virtual try-on methods can only transfer clothes given a fixed human pose, and still show unsatisfactory performances, often failing to preserve person identity or texture details, and with limited pose diversity. This paper makes the first attempt towards a multi-pose guided virtual try-on system, which enables clothes to transfer onto a person with diverse poses. Given an input person image, a desired clothes image, and a desired pose, the proposed Multi-pose Guided Virtual Try-On Network (MG-VTON) generates a new person image after fitting the desired clothes into the person and manipulating the pose. MG-VTON is constructed with three stages: 1) a conditional human parsing network is proposed that matches both the desired pose and the desired clothes shape; 2) a deep Warping Generative Adversarial Network (Warp-GAN) that warps the desired clothes appearance into the synthesized human parsing map and alleviates the misalignment problem between the input human pose and the desired one; 3) a refinement render network recovers the texture details of clothes and removes artifacts, based on multi-pose composition masks. Extensive experiments on commonly-used datasets and our newly-collected largest virtual try-on benchmark demonstrate that our MG-VTON significantly outperforms all state-of-the-art methods both qualitatively and quantitatively, showing promising virtual try-on performances.

关键词： Shape Three-dimensional displays Generators Hair Task analysis Face Gallium nitride

来源：评论

学校读者我要写书评

暂无评论

Corrigendum to “Do fake followers mitigate influencers’ perceived influencing power on social media platforms? The mere number effect and boundary conditions” [J. Bus. Res. 158 (2023) 113589]

引用

Journal of Business Research 2023年 160卷

作者： Liying Zhou Fei Jin Banggang Wu Zhi Chen Cheng Lu Wang School of Business Administration Guizhou University of Finance and Economics PO Box: 550025 Guiyang China Guizhou Key Laboratory of Big Data Statistical Analysis PO Box: 550025 Guiyang China Business School Sichuan Univeristy PO Box: 610065 Guiyang China School of Business Administration Shanghai Lixin University of Accounting and Finance PO Box: 201209 Shanghai China School of Management and Economics North China University of Water Resources and Electric Power Henan China University of New Haven West Haven CT 06516 USA

来源：评论

学校读者我要写书评

暂无评论

FW-GAN: Flow-Navigated Warping GAN for Video Virtual Try-On

FW-GAN: Flow-Navigated Warping GAN for Video Virtual Try-On

引用

International Conference on Computer Vision (ICCV)

作者： Haoye Dong Xiaodan Liang Xiaohui Shen Bowen Wu Bing-Cheng Chen Jian Yin School of Data and Computer Science Sun Yat-sen University Guangdong Key Laboratory of Big Data Analysis and Processing Guangzhou P.R.China School of Intelligent Systems Engineering Sun Yat-sen University ByteDance AI Lab

ISBN: (数字)9781728148038

ISBN: (纸本)9781728148045

Beyond current image-based virtual try-on systems that have attracted increasing attention, we move a step forward to developing a video virtual try-on system that precisely transfers clothes onto the person and generates visually realistic videos conditioned on arbitrary poses. Besides the challenges in image-based virtual try-on (e.g., clothes fidelity, image synthesis), video virtual try-on further requires spatiotemporal consistency. Directly adopting existing image-based approaches often fails to generate coherent video with natural and realistic textures. In this work, we propose Flow-navigated Warping Generative Adversarial Network (FW-GAN), a novel framework that learns to synthesize the video of virtual try-on based on a person image, the desired clothes image, and a series of target poses. FW-GAN aims to synthesize the coherent and natural video while manipulating the pose and clothes. It consists of: (i) a flow-guided fusion module that warps the past frames to assist synthesis, which is also adopted in the discriminator to help enhance the coherence and quality of the synthesized video; (ii) a warping net that is designed to warp clothes image for the refinement of clothes textures; (iii) a parsing constraint loss that alleviates the problem caused by the misalignment of segmentation maps from images with different poses and various clothes. Experiments on our newly collected dataset show that FW-GAN can synthesize high-quality video of virtual try-on and significantly outperforms other methods both qualitatively and quantitatively.

关键词： Integrated circuits Generators Gallium nitride Three-dimensional displays Image generation Computational modeling Visualization

来源：评论

学校读者我要写书评

暂无评论

Evaluation of e-commerce location and entry to China – implications on shipping and trade

引用

Journal of Shipping and Trade 2019年第1期4卷 1-25页

作者： Panova, Yulia Tan, Albert Hilmola, Olli-Pekka Puvindran, M. H. Hongsheng, Xu Li, Weidong Luoyang Normal University Luoyang China Malaysia Institute for Supply Chain Innovation Shah Alam Malaysia LUT University Kouvola Finland School of Economics & Management Beijing Jiaotong University Beijing China Henan Key Laboratory for Big Data Processing & Analytics of Electronic Commerce Luoyang China

E-commerce is experiencing strong global growth, and leading market is nowadays that of China, whether it is evaluated from the perspective of domestic market size or cross-border volumes. In this research work further understanding and knowledge is built from Chinese market using general choice algorithm (GCA) and two real-life case studies. The outcome of GCA model shows the list of the most preferable locations for e-commerce enterprise development. Findings of this study are also compared to two case studies of e-commerce entrance to Chinese market, where one of these two is operating in an omnichannel environment. Market entrance in these two was implemented through very simplified and centralized distribution structure, and using location of Shanghai. Actual operations are either outsourced or in-house produced. Cash on delivery (COD) is still the preferred payment method in e-commerce order fulfilment. One factor being rather important in case companies was distribution prices, where courier services all over the country play an important role. For logistics in general, and sea ports, research reveals that currently dominating areas of the Pearl and Yangtze River Delta face the challenge to some extent from other places of China as favourable locations for e-commerce and consumption are also at north and western parts.

关键词：

来源：评论

学校读者我要写书评

暂无评论

A QoS-based User Reputation Measurement Method for Web Services

A QoS-based User Reputation Measurement Method for Web Servi...

引用

2018 International Conference on Computer Science, Electronics and Communication Engineering（CSECE2018）

作者： Jianlong Xu Yindong Chen Changsheng Zhu Shantou University Hanshan Normal University Guangdong Key Laboratory of Big Data Analysis and Processing

ISBN: (纸本)9781510862593

In Web services environment, high-quality services discovery and selection depend on reliable users with high reputations. However, user reputation is hard to accurately obtain directly. To evaluate the reliability of users, we present user reputation measurement method, named URMM. Based on the historical Quality of Service(QoS) values of users, we calculate user reputation value by the median and iteration operation. Based on real world public datasets, we conduct extensive experiments and demonstrate our proposed approach is more simple and effective than traditional approach.

关键词： Web services quality of service reputation

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：