检索结果-内蒙古大学图书馆

60th Anniversary symposium of the International-Association-for-Shell-and-Spatial-Structures (IASS symposium) / 9th International Conference on Textile Composites and Inflatable Structures (STRUCTURAL MEMBRANES)

作者： Tamura, Tetsuro Cao, Yong Nagao, Yuki Kawai, Hidenori Nishiguchi, Koji Tsubokura, Makoto Tokyo Inst Technol Dept Architecture & Bldg Engn Yokohama Kanagawa 2268502 Japan RIKEN Ctr Computat Sci Wako Saitama Japan

ISBN: (纸本)9788412110104

In this study, high performance computing (HPC) was performed for estimation of wind-induced pressures on various kinds of dome using the K computer (Japanese supercomputer). Focusing on the super critical Reynolds number (Re) and complexity of a long span roof of actual open stadium, real conditions are set up for discussing aerodynamic characteristics of a dome. Using the LES model for strong wind, this study numerically elucidates wind flow patterns and unsteady pressures around a dome with various shapes and provides the information of wind load acting on a dome under the real conditions for a wind resistant design.

关键词： wind engineering high-performance computing

来源：评论

学校读者我要写书评

暂无评论

A Multimodal high Level Video Segmentation for Content Targeted Online Advertising 15th

A Multimodal High Level Video Segmentation for Content Targe...

引用

15th International symposium on Visual computing, ISVC 2020

作者： Mocanu, Bogdan Tapu, Ruxandra Zaharia, Titus ARTEMIS Department Institut Polytechnique de Paris Télécom SudParis Laboratoire SAMOVAR 9 rue Charles Fourier Évry91000 France Telecommunication Department Faculty of ETTI University "Politehnica" of Bucharest Splaiul Independentei 313 Bucharest060042 Romania

ISBN: (纸本)9783030645588

In this paper we introduce a novel advertisement system, dedicated to multimedia documents broadcasted over the Internet. the proposed approach takes into account the consumer’s perspective and inserts contextual relevant ads at the level of the scenes boundaries, while reducing the degree of intrusiveness. From the methodological point of view, the major contribution of the paper concerns a temporal video segmentation method into scenes based on a multimodal (visual, audio and semantic) fusion of information. the experimental evaluation, carried out on a large dataset with more than 30 video documents validates the proposed methodology with average F1 scores superior to 85%. © 2020, Springer Nature Switzerland AG.

关键词： Large dataset

来源：评论

学校读者我要写书评

暂无评论

基于FPGA的浮点可分离卷积神经网络加速方法

引用

高技术通讯 2022年第5期32卷 441-453页

作者：张志超王剑章隆兵肖俊华计算机体系结构国家重点实验室(中国科学院计算技术研究所) 北京100190 中国科学院计算技术研究所北京100190 中国科学院大学北京100049 中国电子科技集团公司第十五研究所北京100083

针对可分离卷积神经网络在星载飞机目标型号分类应用中存在的速度瓶颈以及功耗限制等问题,提出了一种基于现场可编程门阵列(FPGA)数据流调度的浮点深度分离卷积神经网络加速方法,对通用MobileNet的图像分类模型进行加速。采用基于乘法... 详细信息

针对可分离卷积神经网络在星载飞机目标型号分类应用中存在的速度瓶颈以及功耗限制等问题,提出了一种基于现场可编程门阵列(FPGA)数据流调度的浮点深度分离卷积神经网络加速方法,对通用MobileNet的图像分类模型进行加速。采用基于乘法矩阵与前向加法树的深度分离卷积计算阵列设计,解决了深度分离卷积浮点加速的线速吞吐瓶颈。实验结果表明,基于FPGA的目标分类速度为633 FPS,功耗为22.226 W,运算性能为236.04 GFLOPS,计算速度达到了Titan Xp GPU的1.10~2.61倍,计算效能是Titan Xp GPU的7.44~18.66倍。在同类基于FPGA的浮点卷积加速方案中,该方法在运算性能及能效比上达到了最优。同时,该方法提供了与原模型一致性的图像分类准确率,解耦合了软硬件协同开发流程,降低了应用开发人员使用FPGA加速计算的门槛。

关键词：深度可分离卷积现场可编程门阵列(FPGA) 数据流调度加速图像分类

来源：评论

学校读者我要写书评

暂无评论

Localization Research Based on Low Cost Sensor 15th

Localization Research Based on Low Cost Sensor

引用

15th International Conference on Green, Pervasive, and Cloud computing, GPC 2020

作者： Zuo, Jian Zhang, Chenghao Shu, Kuang-I Zhang, Heng College of Computer and Information Science Southwest University Chongqing China E/E Architecture Department Beijing New Energy Vehicle Technology Innovation Center Co. Ltd. Beijing China

ISBN: (纸本)9783030642426

With the increasing demand for indoor navigation applications, indoor navigation has become a research hotspot in many technical fields. high-precision sensors are expensive, and they are often used in industrial and aerospace applications. Low-cost sensors can cause severe drift and noise. this paper uses a simple, low-cost Inertial measurement unit (IMU) and effectively evaluates the precise position of pedestrians. this paper designs a pedestrian position estimation based on low-cost inertial sensors. To solve the problem of inertial sensing noise and accumulated errors during the movement, based on the EKF algorithm, a multi-condition threshold detection method is proposed, and then a zero-speed update, an angular speed update, and a yaw angle update are designed. In the measurement state, the proposed measurement error vector effectively eliminates the cumulative error during walking. Finally, using this algorithm to perform straight and rectangular walking routes, corresponding experiments were performed on the effectiveness of the method used. the proposed method for estimating the inertial navigation of pedestrians has an error within 3.8%. this basic research is of great significance in rehabilitation training and somatosensory play. © 2020, Springer Nature Switzerland AG.

关键词： Aerospace applications

来源：评论

学校读者我要写书评

暂无评论

Cloud architecture for Remote Medical Monitoring

Cloud Architecture for Remote Medical Monitoring

引用

International computer Sciences and Information Technologies (CSIT)

作者： Natalia Axak Mykola Korablyov Matvii Ushakov Kharkiv National University of Radio Electronics Kharkiv Ukraine

ISBN: (数字)9781728174433

ISBN: (纸本)9781728174440

the need to create a cloud architecture is due to the fact that most people do not pay proper attention to their health, often putting off an appointment with a doctor indefinitely. this circumstance in some cases can lead to grave consequences. Using the remote cloud monitoring model will allow patients registered in a medical institution to receive medical care in a timely manner due to remote automated health monitoring. the provision of services and emergency decisions with various levels of complexity is realized through the interaction of the web interface, the layer of intelligent data processing, the agent unit, the accumulation and analysis of experience, as well as by adapting the neural network data processing for high-performance computing systems.

关键词： Cloud computing Monitoring Computational modeling Medical services computer architecture Medical diagnostic imaging Adaptation models

来源：评论

学校读者我要写书评

暂无评论

performance Optimization of Cloud Application at Software architecture Level 11th

Performance Optimization of Cloud Application at Software Ar...

引用

11th International symposium on Intelligence Computation and Applications, ISICA 2019

作者： Du, Xin Ni, Youcong Ye, Peng Wang, Xin Xiao, Ruliang College of Mathematics and Informatics Fujian Normal University Fuzhou350117 China Fujian Provincial Engineering Technology Research Center for Public Service Fuzhou350117 China College of Mathematics and Computer Wuhan Textile University Wuhan430200 China

ISBN: (纸本)9789811555763

Due to the limited search space in the existing performance optimization approaches at software architectures of cloud applications (SAoCA) level, it is difficult for these methods to obtain the cloud resource usage scheme with optimal cost-performance ratio. Aiming at this problem, this paper firstly defines a performance optimization model called CAPOM that can enlarge the search space effectively. Secondly, an efficient differential evolutionary optimization algorithm named MODE4CA is proposed to solve the CAPOM model by defining evolutionary operators with strategy pool and repair mechanism. Further, a method for optimizing performance at SAoCA level, called POM4CA is derived. Finally, two problem instances with different sizes are taken to conduct the experiments for comparing POM4CA with the current representative method under the light and heavy workload. the experimental results show that POM4CA method can obtain better response time and spend less cost of cloud resources. © 2020, Springer Nature Singapore Pte Ltd.

关键词： Software architecture

来源：评论

学校读者我要写书评

暂无评论

Where’s Wally: A Gigapixel Image Study for Face Recognition in Crowds 15th

Where’s Wally: A Gigapixel Image Study for Face Recognition...

引用

15th International symposium on Visual computing, ISVC 2020

作者： Ferreira, Cristiane B. R. Pedrini, Helio de Souza Alencar, Wanderley Ferreira, William D. Carvalho, thyago Peres Sousa, Naiane Soares, Fabrizzio Instituto de Informática Universidade Federal de Goiás GoianiaGO Brazil Institute of Computing University of Campinas CampinasSP Brazil Department of Computer Science Southern Oregon University AshlandOR United States

ISBN: (纸本)9783030645588

Several devices are capable of capturing images with a large number of people, including those of high resolution known as gigapixel images. these images can be helpful for studies and investigations, such as finding people in a crowd. Although they can provide more details, the task of identifying someone in the crowd is quite challenging and complex. In this paper, we aim to assist the work of a human observer with larger images with crowds by reducing the search space for several images to a ranking of ten images related to a specific person. Our model collects faces in a crowded gigapixel image and then searches for people using three different poses (front, right and left). We built a handcraft dataset with 42 people to evaluate our method, achieving a recognition rate of 69% in the complete dataset. We highlight that, from the 31% "not found" among the top ten in the ranking, many of them are very close to this boundary and, in addition, 92% of non-matched are occluded by some accessory or another face. Experimental results showed great potential for our method to support a human observer to find people in the crowd, especially cluttered images, providing her/him with a reduced search space. © 2020, Springer Nature Switzerland AG.

关键词： Face recognition

来源：评论

学校读者我要写书评

暂无评论

An Efficient Text Labeling Framework Using Active Learning Model 1

引用

5th International symposium on Intelligent Systems Technologies and Applications, ISTA 2019

作者： Tandra, Sulochana Nautiyal, Akshay Gupta, Deepa Department of Computer Science and Engineering Amrita School of Engineering Amrita Vishwa Vidyapeetham Bengaluru India

ISBN: (数字)9789811539145

ISBN: (纸本)9789811539138

Electronic medical discharge summaries provide a wealth of information. Extracting useful structured information from such unstructured text is challenging. However, supervised machine learning (ML) algorithms can achieve good performance in extracting useful relations between different entities. To use supervised ML techniques, huge annotated datasets are required. Annotating manually is very expensive and time taking due to the requirement of domain experts for annotation. Active learning (AL), a sample selection approach integrated with supervised ML, aims to minimize the annotation cost while maximizing the performance of ML-based models. Active learning leverages the advantage of training the classifier with a limited number of samples but achieving maximum performance. this strategy not only saves time but also decreases the annotation cost involved. Active learning works well with datasets where annotation cost is high, and training a decent classifier with the available annotated dataset is a requirement. the key factor for an active learning model’s success is its selection of samples that needs annotation. the more informative the samples are, the less time it takes to train the supervised model with high accuracy. thus, the query strategy in sample selection plays a vital role in the AL process. In this study, we aim to develop a novel query strategy to select the most informative samples from the dataset that can eventually accelerate the supervised model’s performance. the query strategy is designed using deep reinforcement learning techniques like actor-critic. the performance of the sample selection strategy is determined by finding the accuracy of the model after a predefined number of iterations. © 2020, Springer Nature Singapore Pte Ltd.

关键词： Reinforcement learning

来源：评论

学校读者我要写书评

暂无评论

Gables: A Roofline Model for Mobile SoCs 25

Gables: A Roofline Model for Mobile SoCs

引用

25th IEEE International symposium on high performance computer architecture (HPCA)

作者： Hill, Mark D. Reddi, Vijay Janapa Univ Wisconsin Dept Comp Sci Madison WI 53706 USA Harvard Univ Sch Engn & Appl Sci Cambridge MA 02138 USA

ISBN: (纸本)9781728114446

Over a billion mobile consumer system-on-chip (SoC) chipsets ship each year. Of these, the mobile consumer market undoubtedly involving smartphones has a significant market share. Most modern smartphones comprise of advanced SoC architectures that are made up of multiple cores, GPS, and many different programmable and fixed-function accelerators connected via a complex hierarchy of interconnects with the goal of running a dozen or more critical software usecases under strict power, thermal and energy constraints. the steadily growing complexity of a modern SoC challenges hardware computer architects on how best to do early stage ideation. Late SoC design typically relies on detailed full-system simulation once the hardware is specified and accelerator software is written or ported. However, early-stage SoC design must often select accelerators before a single line of software is written. To help frame SoC thinking and guide early stage mobile SoC design, in this paper we contribute the Gables model that refines and retargets the Roofline model-designed originally for the performance and bandwidth limits of a multicore chip-to model each accelerator on a SoC, to apportion work concurrently among different accelerators (justified by our usecase analysis), and calculate a SoC performance upper bound. We evaluate the Gables model with an existing SoC and develop several extensions that allow Gables to inform early stage mobile SoC design.

关键词： Accelerator architectures Mobile computing Processor architecture System-on-Chip

来源：评论

学校读者我要写书评

暂无评论

Clocking for HPC Design: Challenges and Experience Sharing

Clocking for HPC Design: Challenges and Experience Sharing

引用

International symposium on VLSI Design, Automation and Test (VLSI-DAT)

作者： Chou, Yih-Chih Wu, Chien-Cheng Tsai, Cheng-Hong Global Unichip Corp 10 Li Hsin 6th RdHsinchu Sci Pk Hsinchu 30078 Taiwan

ISBN: (纸本)9781728106557

In recent years, as the progress of VLSI technology, artificial intelligence / deep learning has become a major trend. To satisfy the demand of high performance computing, many AI ASICs adopt multi-core architecture. the challenges for clocking of this architecture consists of timing closure issues and the implementation of low latency, low skew and low OCV top-level clock tree for highspeed operation (> 1GHz). In this paper, we share our experiences on clock tree synthesis of HPC ASICs. Two different clocking strategies are introduced including H-tree planning with customized big drivers and clock mesh.

关键词： Clocks Timing computer architecture Routing high performance computing Very large scale integration Microprocessors

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：