检索结果-内蒙古大学图书馆

arXiv 2022年

作者： Chen, Ran Wang, Hanli Wang, Lei Kwong, Sam The Department of Computer Science & Technology Key Laboratory of Embedded System and Service Computing Ministry of Education Tongji University Shanghai200092 China Frontiers Science Center for Intelligent Autonomous Systems Shanghai201210 China DeepBlue Academy of Sciences Shanghai200336 China The Department of Computer Science City University of Hong Kong Hong Kong

Reasoning-based approaches have demonstrated their powerful ability for the task of image-text matching. In this work, two issues are addressed for image-text matching. First, for reasoning processing, conventional approaches have no ability to find and use multi-level hierarchical similarity information. To solve this problem, a hierarchical similarity reasoning module is proposed to automatically extract context information, which is then co-exploited with local interaction information for efficient reasoning. Second, previous approaches only consider learning single-stream similarity alignment (i.e., image-to-text level or text-to-image level), which is inadequate to fully use similarity information for image-text matching. To address this issue, a two-stream architecture is developed to decompose image-text matching into image-to-text level and text-to-image level similarity computation. These two issues are investigated by a unifying framework that is trained in an end-to-end manner, namely two-stream hierarchical similarity reasoning network. The extensive experiments performed on the two benchmark datasets of MSCOCO and Flickr30K show the superiority of the proposed approach as compared to existing state-of-the-art methods. Copyright © 2022, The Authors. All rights reserved.

关键词： Convolutional neural networks

来源：评论

学校读者我要写书评

暂无评论

MVC-VPR: Mutual Learning of Viewpoint Classification and Visual Place Recognition

arXiv

引用

arXiv 2024年

作者： Gu, Qiwen Wang, Xufei Zhang, Fenglin Zhao, Junqiao Tao, Siyue Ye, Chen Feng, Tiantian Jiang, Changjun Department of Computer Science and Technology School of Electronics and Information Engineering Tongji University Shanghai China The Shanghai Research Institute for Intelligent Autonomous System Tongji University Shanghai China The MOE Key Lab of Embedded System and Service Computing Tongji University Shanghai China School of Surveying and Geo-Informatics Tongji University Shanghai China

Visual Place Recognition (VPR) aims to robustly identify locations by leveraging image retrieval based on descriptors encoded from environmental images. However, drastic appearance changes of images captured from different viewpoints at the same location pose incoherent supervision signals for descriptor learning, which severely hinder the performance of VPR. Previous work proposes classifying images based on manually defined rules or ground truth labels for viewpoints, followed by descriptor training based on the classification results. However, not all datasets have ground truth labels of viewpoints and manually defined rules may be suboptimal, leading to degraded descriptor performance. To address these challenges, we introduce the mutual learning of viewpoint self-classification and VPR. Starting from coarse classification based on geographical coordinates, we progress to finer classification of viewpoints using simple clustering techniques. The dataset is partitioned in an unsupervised manner while simultaneously training a descriptor extractor for place recognition. Experimental results show that this approach almost perfectly partitions the dataset based on viewpoints, thus achieving mutually reinforcing effects. Our method even excels state-of-the-art (SOTA) methods that partition datasets using ground truth labels. Copyright © 2024, The Authors. All rights reserved.

关键词： Self-supervised learning

来源：评论

学校读者我要写书评

暂无评论

Facial StO2-based Stress Recognition using Automatic Graph Generation and Dual-Stream GNN

Facial StO2-based Stress Recognition using Automatic Graph G...

引用

IEEE International Conference on Bioinformatics and Biomedicine (BIBM)

作者： Yingjuan Jia Hanpu Wang Xinyu Liu Tong Chen College of Electronic and Information Engineering Southwest University China Chongqing Key Laboratory of Generic Technology and System of Service Robots Southwest University China Institute of Legal Psychology and Intelligent Computing Southwest University China

ISBN: (数字)9798350386226

ISBN: (纸本)9798350386233

The recognition of stress responses in health, resilience, and psychopathology holds significant scientific importance. The biological information carried by the facial tissue oxygen saturation (StO2) is a useful indicator for stress recognition. Although graph-based methods have achieved state-of-the-art (SOTA), there is room for improvement. This study proposes an automatic graph generation method for facial StO2, replacing manual methods and enhancing the information of graphs. To further enhance the capability of graphs’ representation, a dimensionality reduction strategy for nodes is proposed. The strategy is implemented based on an novel objective function that can increase the inter-class distance and reduce the intra-class distance. After obtaining highly representative graphs, a dual-stream network integrating GAT and GCN is designed to extract stress-related features from the graphs, ultimately achieving the SOTA recognition accuracy.

关键词： Dimensionality reduction Accuracy Face recognition Manuals Feature extraction Linear programming Biology Convolutional neural networks Stress Resilience

来源：评论

学校读者我要写书评

暂无评论

Asymptotic throughput for large-scale wireless networks with general node density

引用

Wireless Networks 2012年第5期19卷 559-575页

作者： Cheng Wang Changjun Jiang Xiang-Yang Li Yunhao Liu Department of Computer Science Tongji University Shanghai China Key Laboratory of Embedded System and Service Computing Ministry of Education Shanghai China The Tsinghua National Laboratory for Information Science and Technology (TNLIST) Department of Computer Science and Engineering Tongji University Shanghai China The Department of Computer Science Illinois Institute of Technology Chicago USA The TNLIST and School of Software Tsinghua University Beijing China CSE The Hong Kong University of Science and Technology Kowloon Hong Kong

We study the asymptotic throughput for a large-scale wireless ad hoc network consisting of n nodes under the generalized physical model. We directly compute the throughput of multicast sessions to unify the unicast and broadcast throughputs. We design two multicast schemes based on the so-called ordinary arterial road system and parallel arterial road system, respectively. Correspondingly, we derive the achievable multicast throughput by taking account of all possible cases of n s = ω(1) and 1 ≤ n d ≤ n − 1, rather than only the cases of $$n_s=\Uptheta(n)$$ as in most related works, where n s and n d denote the number of sessions and the number of destinations of each session, respectively. Furthermore, we consider the network with a general node density $$\lambda \in [1,n]$$ , while the models in most related works are either random dense network (RDN) or random extended network (REN) where the density is λ = n and λ = 1, respectively, which further enhances the generality of this work. Particularly, for the special case of our results by letting λ = 1 and $$n_s=\Uptheta(n)$$ , we show that for some regimes of n d , the multicast throughputs achieved by our schemes are better than those derived by the well-known percolation-based schemes.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Optimal cooperative relay deployment strategy in heterogeneous cellular network

引用

Journal of Information and Computational Science 2011年第8期8卷 1457-1470页

作者： Gao, Zhiwei Wang, Ping Liu, Fuqiang Shan, Lianhai Key Laboratory of Embedded Systems and Service Computing Ministry of Education of China College of Electronics and Information Engineering Tongji University Shanghai 201804 China Shanghai Research Center for Wireless Communications 280 Linhong Rd Changning District Shanghai 200335 China Shanghai Key Laboratory of Digital Media Processing and Transmission 800 Dongchuan RD. Minhang District Shanghai 201100 China

For the purpose of reducing energy consumption and increasing spatial reuse, relay node deployment strategies have been studied widely in wireless network planning. In this paper, we propose a new deploy strategy when cooperative relay transmission protocol is adopted in the heterogeneous network environment. The existence of cooperative relay nodes introduces more interference in the cellular domain. First, we introduce the hetero-interference models which have great influence on heterogeneous access performance. Then, outage probability and symbol error rate models are derived for heterogeneous relay based network. On this basis, several relay deployment strategies are introduced to analyze cooperative network. Furthermore, the multi-objective solution is used to optimize cooperative relay deployment strategy. Numerical results show that such kind deployment strategy can improve the cellular outage probability and symbol error rate as compared with fixed deployment strategy and Nearest-Neighbor algorithm. © 2011 Binary Information Press.

关键词： Multiobjective optimization

来源：评论

学校读者我要写书评

暂无评论

Model checking cooperative multi-agent systems in BDI logic

引用

Journal of Information and Computational Science 2012年第5期9卷 1185-1194页

作者： Chen, Qingliang Su, Kaile Wu, Lijun Xu, Zhaocheng School of Electronics Engineering and Computer Science Peking University Beijing 100871 China Department of Computer Science Jinan University Guangzhou 510632 China Key Laboratory of Embedded System and Service Computing Ministry of Education Tongji University Shanghai China Institute for Integrated and Intelligent Systems Griffith University Brisbane Australia School of Computer Science and Engineering University of Electronic Science and Technology of China Chengdu China

Traditional temporal logics such as LTL (Linear Temporal Logic) and CTL (Computation Tree Logic) have shown tremendous success in specifying and verifying hardware and software systems. However, this kind of logic can only model the dynamic behaviors for a single system, and fall short in capturing the joint and concurrent aspects in the composition of systems, such as the interactions and coordinations of agents in Multi-agent systems. Although ATL (Alternating-time Temporal Logic) can solve the problem to a certain extent, by preceding the temporal operators with a selective path quantifier, to specify that the interactions between entities inside the system can make sure that there is a strategy to reach a certain state, it can not formalize the rational mental attitudes of the agents such as the Belief, Desire and Intention (BDI) which plays an essential role in decision making during cooperation with other agents. This paper will investigate this problem by proposing a logic of ATLBDI (Alternating-time Temporal Logic of Belief Desire and Intention) and discuss the algorithmic issues such as model checking algorithms and the computational complexity, along with examples to show its effectiveness. We conclude that model checking for ATLBDI is completely tractable and is in PTIME-Complete, which is quite an optimistic and promising result for further applications. © 2012 by Binary Information Press.

关键词： Multi agent systems

来源：评论

学校读者我要写书评

暂无评论

arXiv

引用

arXiv 2017年

作者： Lang, Guangming School of Mathematics and Statistics Changsha University of Science and Technology Changsha Hunan410114 China Department of Computer Science and Technology Tongji University Shanghai201804 China The Key Laboratory of Embedded System and Service Computing Ministry of Education Tongji University Shanghai201804 China College of Mathematics and Econometrics Hunan University Changsha Hunan410004 China

In practice, there are many dynamic covering decision information systems, and knowledge reduction of dynamic covering decision information systems is a significant challenge of covering-based rough sets. In this paper, we first study mechanisms of constructing attribute reducts for consistent covering decision information systems when adding objects using related families. We also employ examples to illustrate how to construct attribute reducts of consistent covering decision information systems when adding objects. Then we investigate mechanisms of constructing attribute reducts for consistent covering decision information systems when deleting objects using related families. We also employ examples to illustrate how to construct attribute reducts of consistent covering decision information systems when deleting objects. Finally, the experimental results illustrates that the related family-based methods are effective to perform attribute reduction of dynamic covering decision information systems when object sets are varying with time. Copyright © 2017, The Authors. All rights reserved.

关键词： Rough set theory

来源：评论

学校读者我要写书评

暂无评论

arXiv

引用

arXiv 2017年

作者： Lang, Guangming School of Mathematics and Statistics Changsha University of Science and Technology Changsha Hunan410114 China Department of Computer Science and Technology Tongji University Shanghai201804 China Key Laboratory of Embedded System and Service Computing Ministry of Education Tongji University Shanghai201804 China College of Mathematics and Econometrics Hunan University Changsha Hunan410004 China

In practical situations, there are many dynamic covering information systems with variations of attributes, but there are few studies on related family-based attribute reduction of dynamic covering information systems. In this paper, we first investigate updated mechanisms of constructing attribute reducts for consistent and inconsistent covering information systems when varying attribute sets by using related families. Then we employ examples to illustrate how to compute attribute reducts of dynamic covering information systems with variations of attribute sets. Finally, the experimental results illustrates that the related family-based methods are effective to perform attribute reduction of dynamic covering information systems when attribute sets are varying with time. Copyright © 2017, The Authors. All rights reserved.

关键词： Information use

来源：评论

学校读者我要写书评

暂无评论

On iterative liveness-enforcement for a class of generalized Petri nets

On iterative liveness-enforcement for a class of generalized...

引用

IEEE International Conference on Automation Science and Engineering (CASE)

作者： YiFan Hou Ding Liu MengChu Zhou School of Electro-Mechanical Engineering Xidian University Xian China Key Laboratory of Embedded System and Service Computer Ministry of Education Tongji University Shanghai China Department of Electrical and Computer Engineering New Jersey Institute of Technology Newark NJ USA

This work presents an iterative liveness-enforcing method for a class of generalized Petri nets, which can model flexible manufacturing systems. The proposed method checks the liveness of net models using mixed integer programming and controls the token allocations of resource places instead of siphons using a liveness and resource usage ratio-enforcing supervisor. The enumeration of a kind of special structures, which is required in the previous work, is avoided and the number of iterations is bounded by the number of shared resource places in the net model. All strict minimal siphons in the controlled systems are minimally controlled. Several explanatory examples are used to illustrate this method.

关键词： Monitoring Petri nets Iterative methods system recovery Educational institutions Resource management Vectors

来源：评论

学校读者我要写书评

暂无评论

A novel visible-infrared image fusion framework for smart city

引用

International Journal of Simulation and Process Modelling 2018年第2期13卷 144-155页

作者： Zhu, Zhiqin Qi, Guanqiu Chai, Yi Yin, Hongpeng Sun, Jian School of Automation Chongqing University Chongqing400044 China State Key Laboratory of Power Transmission Equipment and System Security and New Technology College of Automation Chongqing University Chongqing400044 China School of Computing Informatics and Decision Systems Engineering Arizona State University TempeAZ85287 United States Key Laboratory of Dependable Service Computing in Cyber Physical Society Ministry of Education Chongqing400030 China

Image fusion technology is widely used in different areas and can integrate complementary and relevant information of source images captured by multiple sensors into a unitary synthetic image. Image fusion technology as an efficient way to integrate information from multiple images plays a more and more important role in smart city. The quality of fused image affects the accuracy, efficiency, and robustness of the related applications. Existing sparse representation-based image fusion methods consist of overly complete and redundant dictionaries learning and sparse coding. However, overly complete and redundant dictionary does not consider the discriminative ability of dictionaries that may seriously affect the image fusion. A good dictionary is the key to a successful image fusion technique. To construct a discriminative dictionary, a novel framework that integrates an image-patches clustering and online dictionary learning methods is proposed for visible-infrared image fusion. The comparison experiments with existing solutions are used to validate and demonstrate the effectiveness of the proposed solution for image fusion. Copyright © 2018 Inderscience Enterprises Ltd.

关键词： Image fusion

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：