检索结果-内蒙古大学图书馆

Conference on Computer Vision and Pattern recognition (CVPr)

作者： Xinyu Sun Peihao Chen Liangwei Chen Changhao Li Thomas H. Li Mingkui Tan Chuang Gan South China University of Technology Information Technology R&D Innovation Center of Peking University Peking University Shenzhen Graduate School Key Laboratory of Big Data and Intelligent Robot Ministry of Education Pazhou Laboratory UMass Amherst MIT-IBM Watson AI Lab

How to learn discriminative video representation from unlabeled videos is challenging but crucial for video analysis. The latest attempts seek to learn a representation model by predicting the appearance contents in the masked regions. However, simply masking and recovering appearance contents may not be sufficient to model temporal clues as the appearance contents can be easily reconstructed from a single frame. To overcome this limitation, we present Masked Motion Encoding (MME), a new pretraining paradigm that reconstructs both appearance and motion information to explore temporal clues. In MME, we focus on addressing two critical challenges to improve the representation performance: 1) how to well represent the possible long-term motion across multiple frames; and 2) how to obtain fine-grained temporal clues from sparsely sampled videos. Motivated by the fact that human is able to recognize an action by tracking objects' position changes and shape changes, we propose to reconstruct a motion trajectory that represents these two kinds of change in the masked regions. Besides, given the sparse video input, we enforce the model to reconstruct dense motion trajectories in both spatial and temporal dimensions. Pre-trained with our MME paradigm, the model is able to anticipate long-term and fine-grained motion details. Code is available at https://***/XinyuSun/MME.

关键词：

来源：评论

学校读者我要写书评

暂无评论

redefining Safety Nets: Text detection solution for Cyberbullying Eradication

Redefining Safety Nets: Text Detection solution for Cyberbul...

引用

distributed Systems, Computer Networks and Cybersecurity (ICdSCNC), International Conference on

作者： K. Biksheswara rao B. Suvarnamukhi S. Famila Shanmugasundaram Hariharan Thava Vinu A S.V. Vasantha Department of CSE (Data Science) CMR Technical Campus Hyderabad India Department of CSE Sreyas Institute of Engineering and Technology Hyderabad India Department of ECE Vel Tech Rangarajan Dr.Sagunthala R&D Institute of Science and Technology Chennai India Department of Artificial Intelligence and Data Science Vardhaman College of Engineering Hyderabad India Department of Information Technology Prathyusha Engineering College Thiruvallur India Department of CSE Vardhaman College of Engineering Hyderabad India

ISBN: (数字)9798350375442

ISBN: (纸本)9798350375459

Cyberbullying, the malicious use of digital communication channels to intimidate, harass, or harm individuals, has emerged as a significant concern in the era of widespread internet usage and social media platforms. The anonymous nature of online interactions makes it easier for individuals to engage in abusive behavior, leaving victims vulnerable to emotional distress and potential psychological harm. To address this growing issue, the development of robust and efficient text detection methods for cyberbullying has become imperative. methods, contextual analysis, and user behavior analysis as potential techniques for identifying cyberbullying instances. Each method’s strengths and limitations are discussed, emphasizing the importance of combining multiple techniques to enhance detection accuracy. The study underlines the significance of context-aware analysis in distinguishing between genuine cyberbullying and harm- less banter among friends. Additionally, ethical considerations, user privacy, and transparency in monitoring mechanisms are addressed to ensure a balanced approach to text detection. Finally, the abstract emphasizes the necessity of human review in the detection process to prevent false positives and negatives, fostering a more inclusive and supportive online environment. By leveraging the power of advanced natural language processing techniques, this research contributes to the ongoing efforts to combat cyberbullying effectively and safeguard the emotional well-being of internet users, particularly the most vulnerable among them

关键词： Support vector machines Ethics Privacy Accuracy Cyberbullying Text detection Psychology Machine learning Natural language processing Safety

来源：评论

学校读者我要写书评

暂无评论

An improved energy management model for IoT devices using deep learning

An improved energy management model for IoT devices using de...

引用

International Conference on Computing and Networking technology (ICCNT)

作者： P. roy Sudha reetha S Joseph N. Pandeeswari r Anitha S. Jebarose Juliyana Department of Artificial Intelligence and Data Science PSNA College of Engineering and Technology Dindigul Dt Tamil Nadu India Department of Electronics and Communication Christian College of Engineering and Technology Dindigul Dt Tamil Nadu India Department of Information Technology PSNA College of Engineering and Technology Dindigul Dt Tamil Nadu India Department of Artificial intelligence and Data Science Dr. Mahalingam college of engineering and Technology Pollachi Tamil Nadu India Department of Mechanical Engineering Vel Tech Rangarajan Dr Sagunthala R&D Institute of Science and Technology Chennai Tamil Ndu India

ISBN: (数字)9798350370249

ISBN: (纸本)9798350370270

The Internet of Things (IoT) is booming, with research showing that billions of connected devices gather data, communicate, and automate. The widespread adoption of IoT devices has been hampered by one major challenge: limited energy resources that require recharging or replacing often. Energy management strategies on these devices are essential to improve battery life. This paper presents an enhanced version of the energy management model for IoT devices using deep learning techniques in technical abstracts. The aim is to reduce energy consumption by predicting and configuring real-time device settings based on user behavior trends. A deep neural network is used as the model to learn and tailor its response variably based on different user conditions like temperature, light, and usage patterns. The model can analyze this data to predict energy demand and programmatically adjust device settings.

关键词： deep learning Energy consumption Adaptation models runtime Computational modeling Predictive models Software real-time systems Internet of Things Energy management

来源：评论

学校读者我要写书评

暂无评论

Smartphone-based Strep Throat detection using MTCNN: A deep Learning Approach for Non-Invasive diagnosis

Smartphone-based Strep Throat Detection using MTCNN: A Deep ...

引用

International Conference on I-SMAC (IoT in Social, Mobile, Analytics and Cloud) (I-SMAC)

作者： G. M. Sathyaseelan L. Sankara Maheswari S. Sophiya M. Malini S. P. Santhoshkumar d. Vikram Department of Information Technology JKK Natraja College of Engineering and Technology Kumarapalayam India Department of Information Technology Sri G.V.G Visalakshi College for Women Udumalpet India Department of Computer Science and Engineering Vel Tech Rangarajan Dr. Sagunthala R&D Institute of Science and Technology Avadi Chennai Department of Artificial Intelligence and Data Science Rathinam Technical Campus Coimbatore India

ISBN: (数字)9798350376425

ISBN: (纸本)9798350376432

This research proposes a novel approach for strep throat detection utilizing the Multi-Task Cascaded Convolutional Neural Network (MTCNN) algorithm and a smartphone. By incorporating advanced facial detection techniques, MTCNN enhances the accuracy of throat image analysis. The proposed method involves capturing throat images using a smartphone camera with an added device to reduce light reflection. MTCNN is then applied to extract relevant information from the images, enabling a classification model to distinguish between healthy and strep-infected throats. Experimental results demonstrate the effectiveness of this approach, providing a non-invasive and efficient method for strep throat diagnosis. The integration of MTCNN holds significant potential for smartphone-based healthcare applications due to its high accuracy and enhanced feature extraction capabilities.

关键词： deep learning Accuracy Sensitivity Image analysis Medical services Multitasking Feature extraction reflection Face detection Convolutional neural networks

来源：评论

学校读者我要写书评

暂无评论

Weakly-supervised multi-granularity map learning for vision-and-language navigation 22

Weakly-supervised multi-granularity map learning for vision-...

引用

Proceedings of the 36th International Conference on Neural Information Processing Systems

作者： Peihao Chen dongyu Ji Kunyang Lin runhao Zeng Thomas H. Li Mingkui Tan Chuang Gan South China University of Technology and Information Technology R&D Innovation Center of Peking University South China University of Technology South China University of Technology and Pazhou Laboratory Shenzhen University Information Technology R&D Innovation Center of Peking University South China University of Technology and Key Laboratory of Big Data and Intelligent Robot Ministry of Education MIT-IBM Watson AI Lab and UMass Amherst

ISBN: (纸本)9781713871088

We address a practical yet challenging problem of training robot agents to navigate in an environment following a path described by some language instructions. The instructions often contain descriptions of objects in the environment. To achieve accurate and efficient navigation, it is critical to build a map that accurately represents both spatial location and the semantic information of the environment objects. However, enabling a robot to build a map that well represents the environment is extremely challenging as the environment often involves diverse objects with various attributes. In this paper, we propose a multi-granularity map, which contains both object fine-grained details (e.g., color, texture) and semantic classes, to represent objects more comprehensively. Moreover, we propose a weakly-supervised auxiliary task, which requires the agent to localize instruction-relevant objects on the map. Through this task, the agent not only learns to localize the instruction-relevant objects for navigation but also is encouraged to learn a better map representation that reveals object information. We then feed the learned map and instruction to a waypoint predictor to determine the next navigation goal. Experimental results show our method outperforms the state-of-the-art by 4.0% and 4.6% w.r.t. success rate both in seen and unseen environments, respectively on VLN-CE dataset.

关键词：

来源：评论

学校读者我要写书评

暂无评论

BAA: A Novel decentralized Authorization System for Privacy-Sensitive Medical data

BAA: A Novel Decentralized Authorization System for Privacy-...

引用

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

作者： Cong Zha Yulei Wu Zexun Jiang Wenqian Zhao Hao Yin The Network and Big Data Technology R&D Center Tsinghua University Beijing China The Department of Computer Science Faculty of Environment Exeter University Exeter UK School of Data Science and Intelligent Media Communication University of China Beijing China Department of Information Science University of North Texas Texas United States

ISBN: (数字)9798350381993

ISBN: (纸本)9798350382006

data authorization is the basis for the orderly sharing of medical data. Most of the applied decentralized authorization mechanisms rely on blockchain, but face the problems of privacy leakage and low efficiency. To solve these problems, we design a policy-driven decentralized authorization system named BAA, which protects user's behavior privacy in medical data sharing and improves efficiency of both on-chain and off-chain. In order to achieve these goals, BAA uses authorization tokens to represent permissions, protects privacy of the authorization process through hiding user behaviors, realizes batch data accessing by proposing a two-tier Merkle tree, and saves authorization data in a two-tier blockchain to improve on-chain efficiency. Extensive experimental results show that the overhead of cryptographic operations in BAA is acceptable compared to that in traditional authorization systems. In addition, throughput and latency of each operation in BAA can meet the efficiency needs of medical data authorization. The results also show that the preset authorization in blockchain is effective for reducing data user's waiting time and improving the efficiency of authorization.

关键词： Authorization data privacy Privacy Throughput Blockchains Protection Faces

来源：评论

学校读者我要写书评

暂无评论

A decision-dependent Chance-Constrained Planning Model for distribution Networks Under Extreme Weather Events

A Decision-Dependent Chance-Constrained Planning Model for D...

引用

IEEE PES Innovative Smart Grid Technologies Conference Europe (ISGT Europe)

作者： Anping Zhou Hongbo Sun Shoichi Kitamura daniel Nikovski Dept. of Electrical and Computer Engineering Southern Methodist University Dallas TX USA Advanced Technology R&D Center Mitsubishi Electric Corporation Amagasaki Japan Data Analytics Group Mitsubishi Electric Research Laboratories Cambridge MA USA

Extreme weather events have posed tremendous challenges to the operation of distribution networks. In this paper, we propose a decision-dependent chance-constrained model for the optimal planning of diesel generators, renewable distributed generations (rdGs), energy storage systems, and switches under contingency. A promising moment-based ambiguity set that incorporates the information of decision variables is employed to depict the uncertainty arising from rdGs. By leveraging effective approximation methods such as the Bonferroni approximation method to handle the considered joint chance constraints, the proposed model is transformed into a tractable mixed-integer second-order conic programming problem, which means it can easily be implemented. Numerical experiments are put forward on the IEEE 33-bus test system to validate the effectiveness of the developed approach.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Control Strategy for Cooperative driving Automation in Traffic Incidents

Control Strategy for Cooperative Driving Automation in Traff...

引用

International Conference on Control, Automation and diagnosis (ICCAd)

作者： Sujin Kwag Byeongju Kang Hyeonsoo Jang Junyeop Park Yunhyoung Hwang Department of Big Data and SW Platform R&D Korea Automotive Technology Institute Seongnam-si Gyeonggi-do Rep. of Korea Department of Mechanical Engineering Jeonbuk National University Jeonju-si Jeollabuk-do Rep. of Korea

With the advancement on automated vehicle technology, the highly automated driving is expected to work in complicated traffic situations such as urban intersections and traffic incidents. However, the traffic incidents still remain a comprehensively challenging situation especially in the mixed traffic. To resolve this problem, a sophisticated control strategy is required to support the automated driving in these unexpected traffic situations. From this background, this study mainly focuses on developing a generalized control strategy to manage the automated vehicles driving in traffic incident situations. First, the representative scenarios are analyzed, and a generalized control strategy is elaborated based on the framework of cooperative driving automation. The proposed control strategy can be extended to more complicated situations, and it is expected to extend the operational design domain of automated vehicles.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Estimating infrasonic noise levels using a high resolution weather model

引用

The Journal of the Acoustical Society of America 2024年第3_SUPPLEMENT期155卷 A70-A70页

作者： Jelle d. Assink Madelon Smink Fred Bosveld Läslo G. Evers R&D Seismology and Acoust. KNMI Utrechtseweg 297 Utrecht 3731GA Netherlands jelle.assink@knmi.nl R&D Seismology and Acoust. KNMI De Bilt Netherlands R&D Observations & Data Technol. KNMI De Bilt Netherlands R&D Seismology and Acoust. KNMI Utrecht Netherlands

The presence of turbulence and other wind-induced pressure fluctuations are considered a nuisance in infrasound monitoring. For this reason, wind noise filters are typically in place for suppression. In this study, we establish a relation between microbarometer observations and in situ turbulence measurements at the Cabauw Atmospheric research Site in The Netherlands. Using this relation, we forecast turbulent pressure levels using forecasted levels of turbulence kinetic energy (TKE) that are computed as part of the high-resolution (2.5 × 2.5 km grid scale) HArMONIE weather model. The estimates are compared with pressure noise levels as a function of frequency. This approach has two foreseen applications: (1) the forecasted turbulence fields could possibly help in site selection for infrasound arrays and (2) microbarometer observations could possibly be of use in the further refining of sub-grid scale turbulence schemes in weather models.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Unified Accelerator for Attention and Convolution in Inference Based on FPGA

Unified Accelerator for Attention and Convolution in Inferen...

引用

IEEE International Symposium on Circuits and Systems (ISCAS)

作者： Tianyang Li Fan Zhang Xitian Fan Jianliang Shen Wei Guo Wei Cao Information Engineering University Zhengzhou China National Digital Switching System Engineering and Technological R&D Center Zhengzhou China Shanghai HONGZHEN Information Science & Technology Corporation Shanghai China Institute for Big Data Fudan University Shanghai China

Many models combining Transformers with convolutional neural networks (CNNs) for computer vision tasks have achieved state-of-the-art results. However, due to the different computation patterns between attention and convolution, using a dedicated Transformer or CNN accelerator will inevitably reduce the computing efficiency of the other. To overcome this problem, we propose a unified architecture for attention and convolution on FPGA. We reduce runtime overhead by offloading part of self-attention computations offline before inference. Furthermore, we present a unified mapping method according to the computing characteristics of attention-based and convolution-based models. This accelerator implements multi-head attention in Transformer, independent resNet-50 and hybrid blocks of attention and con-volution in BoTNet-50 at 200MHz on Xilinx Virtex Ultrascale+ XCVU37P. Experimental results show that the solution is nearly 3.62 times more energy-efficient than the NVIdIA V100 GPU, and the computational efficiency is 11.86% and 28.29% higher than the state-of-the-art Transformer and resNet-50 accelerators, respectively.

关键词：

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：