检索结果-内蒙古大学图书馆

CAT: A Simple yet Effective Cross-Attention Transformer for One-Shot Object Detection

Journal of Computer Science & technology 2024年第2期39卷 460-471页

作者：林蔚东邓玉岩高扬王宁刘凌峤张磊王鹏 School of Computer Science Northwestern Polytechnical UniversityXi’an710000China National Engineering Laboratory for Integrated Aero-Space-Ground-Ocean Big Data Application Technology Northwestern Polytechnical UniversityXi'an710000China School of Computer Science The University of AdelaideAdelaideSA 0115Australia

Given a query patch from a novel class,one-shot object detection aims to detect all instances of this class in a target image through the semantic similarity ***,due to the extremely limited guidance in the novel class as well as the unseen appearance difference between the query and target instances,it is difficult to appropriately exploit their semantic similarity and generalize *** mitigate this problem,we present a universal Cross-Attention Transformer(CAT)module for accurate and efficient semantic similarity comparison in one-shot object *** proposed CAT utilizes the transformer mechanism to comprehensively capture bi-directional correspondence between any paired pixels from the query and the target image,which empowers us to sufficiently exploit their semantic characteristics for accurate similarity *** addition,the proposed CAT enables feature dimensionality compression for inference speedup without performance *** experiments on three object detection datasets MS-COCO,PASCAL VOC and FSOD under the one-shot setting demonstrate the effectiveness and efficiency of our model,e.g.,it surpasses CoAE,a major baseline in this task,by 1.0%in average precision(AP)on MS-COCO and runs nearly 2.5 times faster.

关键词： one-shot object detection Transformer attention mechanism

来源：评论

学校读者我要写书评

暂无评论

Real-time Depth Estimation for Aerial Panoramas in Virtual Reality

Real-time Depth Estimation for Aerial Panoramas in Virtual R...

引用

Virtual Reality and 3D User Interfaces Abstracts and Workshops (VRW), IEEE Conference on

作者： Di Xu Xiaojun Liu Yanning Zhang National Engineering Laboratory for Integrated Aero-Space-Ground-Ocean Big Data Application Technology School of Computer Science Northwestern Polytechnical University China

ISBN: (数字)9781728165325

ISBN: (纸本)9781728165332

With the emergence of consumer-level 360° panoramic cameras, omnidirectional RGB images and videos are now easy to be captured either in the hand-held mode or from a drone. Although previous works achieve plausible results in room-sized panoramic datasets, they are limited in indoor scenes with little rotations. In this paper, we present a real-time depth estimation method for more challenging aerial panoramas, where the viewing angle is changing rapidly and the lighting condition is more complicated. Our graph convolutional network(GCN)-based framework makes full use of the global connection information of the omnidirectional images, and is trained with extensive outdoor data. Experiments show that our method is robust to estimate the depth of outdoor aerial panoramas captured from various angles accurately.

关键词： Estimation Videos Distortion Cameras Three-dimensional displays Real-time systems Drones

来源：评论

学校读者我要写书评

暂无评论

An auto code generator for stencil on SW26010 21

An auto code generator for stencil on SW26010

引用

21st IEEE International Conference on High Performance Computing and Communications, 17th IEEE International Conference on Smart City and 5th IEEE International Conference on data Science and Systems, HPCC/SmartCity/DSS 2019

作者： Zhu, Xiaomin Zeng, Yunhui Wei, Yanjie Feng, Shengzhong Liu, Weiguo Balaji, Pavan Joint Engineering Research Center Health Big Data Intelligent Analysis Technology Center for High Performance Computing Shenzhen Institutes of Advanced Technology Chinese Academy of Sciences Shenzhen China Shandong Computer Science Center National SuperComputer Center in Jinan Jinan China Sclool of Software Shandong University Jinan China Argonne National Laboratory United States

ISBN: (纸本)9781728120584

Stencil is a basic building block widely used in many HPC areas and applications. It generally dominates the time cost and is critical to the overall performance. Given that heterogeneous many-core is frequently adopted to build super computers, porting and optimizing stencil codes on modern accelerator-based architectures is important. Porting is non-trivial, and optimization is more difficult as it requires better understanding of the underlying architecture. As a result, auto tuning targeting on accelerators such as GPU becomes a hot research topic. In this paper, we focus on tuning stencil automatically on SW26010, which is a heterogeneous many-core CPU equipped in Sunway TaihuLight SuperComputer, and mainly work on 2D stencil variants code in ROMS (Regional Ocean Modeling System). The code generator can generate codes running on slave cores for 50+ different FORTRAN stencil-alike loops in ROMS in seconds, and optimization methods are integrated into the generator as well. The performance is better than the directive based method (OpenACC) and the hand-drafted code by a junior programmer, and it is comparable compared with the code written by a senior programmer with several years of programming experience. © 2019 IEEE.

关键词： Supercomputers

来源：评论

学校读者我要写书评

暂无评论

Green and Efficient Method for Recycling Valuable Metals from Scrapped Lithium Cobalt Oxide Cathode Materials

Green and Efficient Method for Recycling Valuable Metals fro...

引用

作者： Miaomiao Guo Xiaoli Xi Dongdong Zhang Hang Yang Fan Bi Yi Hu Zuoren Nie Kaihua Xu National Engineering Laboratory for Industrial Big-data Application Technology Beijing University of Technology Chinese National WEEE Recycling Engineering Research Center Jingmen GEM New Material Co. Ltd

In this study, green and efficient method for recycling valuable metals from scrapped lithium cobalt oxide cathode materials to form lithium carbonate and cobalt powder was proposed. To this end, aluminum and iron impurities were efficiently removed from scrapped lithium cobalt oxide cathode materials through alkali dissolution and magnetic separation with removal rates of 99.9% and 98.3%, respectively. After removal of impurities, lithium cobalt oxides were reduced by hydrogen at high temperature of 650 ℃ for 3 h to yield Li2O, Co, and very small amounts of CoO and LiOH. Next, the mixture was subjected to water leaching and sodium carbonate precipitation to obtain lithium carbonate with purity of 99.5%. The waterleaching Co and small amounts of cobalt oxides were again reduced by hydrogen to form cobalt powder with purity of 99.4%. In sum, the proposed method looks promising in terms of efficiency, easy for implementation, environmentally-friendliness, and potential industrialization.

关键词： CoO Green and Efficient Method for Recycling Valuable Metals from Scrapped Lithium Cobalt Oxide Cathode Materials

来源：评论

学校读者我要写书评

暂无评论

An Empirical Guide for Visualization Consistency in Multiple Coordinated Views

An Empirical Guide for Visualization Consistency in Multiple...

引用

Pacific (formerly Asia-Pacific APVIS) Visualization Symposium

作者： Shaocong Tan Chufan Lai Xiaolong Luke Zhang Xiaoru Yuan Key Laboratory of Machine Perception (Ministry of Education) and School of Intelligence Science and Technology Peking University China College of Information Sciences and Technology Pennsylvania State University USA National Engineering Laboratory for Big Data Analysis and Application Peking University China

Visual analytic systems usually provide multiple coordinated views (MCVs) to support data analysis and exploration. Coordination in visual graphics plays an important role in facilitating comprehensive analytical tasks, such as data comparison and cognitive inference. However, individual views in MCVs are probably designed for a specific purpose based on a particular type of data, and insufficient consideration of the intricate relationships among views may lead to inconsistency in visual representation and user interaction across different views. To better understand the inconsistency issues in MCVs and their impacts on user behaviors, this paper reports a study on the analysis and classification of visualization inconsistency based on the reviews of interactive visualization designs and visual analytic systems, and the interviews with stakeholders. We find that inconsistencies are prevalent in MCVs and frequently lead to misleading or even incorrect results. We classify the discovered inconsistencies based on a coordination model of MCVs, and develop an empirical guide for systematic and efficient visualization consistency checking in the design, implementation, and evaluation stage.

关键词：

来源：评论

学校读者我要写书评

暂无评论

PrompTHis: Visualizing the Process and Influence of Prompt Editing during Text-to-Image Creation

arXiv

引用

arXiv 2024年

作者： Guo, Yuhan Shao, Hanning Liu, Can Xu, Kai Yuan, Xiaoru Key Laboratory of Machine Perception Ministry of Education School of AI Peking University China National Engineering Laboratory for Big Data Analysis and Application Peking University China University of Nottingham United Kingdom

Generative text-to-image models, which allow users to create appealing images through a text prompt, have seen a dramatic increase in popularity in recent years. However, most users have a limited understanding of how such models work and it often requires many trials and errors to achieve satisfactory results. The prompt history contains a wealth of information that could provide users with insights into what have been explored and how the prompt changes impact the output image, yet little research attention has been paid to the visual analysis of such process to support users. We propose the Image Variant Graph, a novel visual representation designed to support comparing prompt-image pairs and exploring the editing history. The Image Variant Graph models prompt differences as edges between corresponding images and presents the distances between images through projection. Based on the graph, we developed the PrompTHis system through co-design with artists. Besides Image Variant Graph, PrompTHis also incorporates a detailed prompt-image history and a navigation mini-map. Based on the review and analysis of the prompting history, users can better understand the impact of prompt changes and have a more effective control of image generation. A quantitative user study with eleven amateur participants and qualitative interviews with five professionals and one amateur user were conducted to evaluate the effectiveness of PrompTHis. The results demonstrate PrompTHis can help users review the prompt history, make sense of the model, and plan their creative process. © 2024, CC BY.

关键词： Image analysis

来源：评论

学校读者我要写书评

暂无评论

Optimal Scheduling of Single-Line Electric Buses with Battery Swapping

Optimal Scheduling of Single-Line Electric Buses with Batter...

引用

IEEE International Conference on Intelligent Transportation engineering (ICITE)

作者： Yuyan Guo Tao Liu National Engineering Laboratory of Integrated Transportation Big Data Application Technology Institute of System Science and Engineering School of Transportation and Logistics Southwest Jiaotong University Chengdu China

ISBN: (数字)9798350314212

ISBN: (纸本)9798350314229

With the advantages of zero emission and comfortable riding experience, battery-electric buses (BEBs) are widely adopted in public transit agencies as a green alternative to conventional diesel buses. Battery swapping technology is an efficient and promising charging technology to address the issues of battery range anxiety and long battery charging time. This study developed a two-stage optimization model to optimize the single-line battery-electric bus scheduling problem considering battery swapping. At the first stage, a bi-objective integer programming model is established to determine the minimum BEB fleet size, together with the BEB-to-trip assignment and battery swapping schedule. At the second stage, an integer linear programming model is formulated to determine the required minimum number of batteries. An $\epsilon$ -constraint method is designed to investigate the trade-off between the required minimum fleet size and the number of back-up batteries. Finally, a real-world case study of a BEB line in Chengdu, China, is conducted to demonstrate the effectiveness of the optimization model and solution method. Sensitivity analyses are further conducted to understand the impacts of some key parameters, including battery capacity, number of trips, cycle time, and battery swapping time. The results show that battery capacity and cycle time have significant influences on BEB-to-trip assignment results and battery swapping schedule.

关键词： Integer programming Schedules Sensitivity analysis Optimization models Transportation Stochastic processes Optimal scheduling Integer linear programming Scheduling Batteries

来源：评论

学校读者我要写书评

暂无评论

Online Public Opinion Deduction Based on an Innovative Cellular Automata 3rd

Online Public Opinion Deduction Based on an Innovative Cellu...

引用

3rd International Conference on Cyberspace data and Intelligence, Cyber DI 2019, and the International Conference on Cyber-Living, Cyber-Syndrome, and Cyber-Health, CyberLife 2019

作者： Liu, Xin Cao, Shuai Cao, Yang He, Jie Zhang, Weishan Wang, Xueli Zheng, Liang College of Computer Science and Technology China University of Petroleum Qingdao266580 China Application on Improving Government Governance Capabilities National Engineering Laboratory Guiyang550022 China CETC Big Data Research Institute Co. Ltd. Guiyang550022 China Information Science Academy of China Electronics Technology Group Corporation Beijing China

ISBN: (纸本)9789811519215

As the Internet entered the 2.0 era, online users have become the main part of public opinion, and significant events happened around the world quickly disseminate through the Internet. Some negative comments usually are mixed into posts and articles. If let public opinion develop at will, some serious social events may trigger very fierce online volatility and they will have adverse effects on social stability. Therefore, to simulate the propagation mode of topic events on the Internet, and to predict the trending of public opinion on the entire Internet, this paper constructs a new type of model based on the idea of cellular automata and applies it to the online public opinion situation deduction. We arrange the nodes in new ways. An individual user is represented by a cell node in our model, and we build the cellular space based on directed graph to represent social network topology that users follow about each other in an innovative way. We delineate the users’ portrait based on the users’ attributes, design evolutionary rules based on the behaviors of users on the Internet and the law of information dissemination. Experiments with real data of online users showed that the proposed model in this paper had excellent performance, it not only reflected the law of information dissemination on the Internet, but also its deduction result was similar to the trending of real historical events. © 2019, Springer Nature Singapore Pte Ltd.

关键词： Cellular automata

来源：评论

学校读者我要写书评

暂无评论

Edit-History Vis: An Interactive Visual Exploration and analysis on Wikipedia Edit History

Edit-History Vis: An Interactive Visual Exploration and Anal...

引用

Pacific (formerly Asia-Pacific APVIS) Visualization Symposium

作者： Yuhan Guo Qin Han Yuke Lou Yiming Wang Can Liu Xiaoru Yuan Key Laboratory of Machine Perception (Ministry of Education) and School of Intelligence Science and Technology Peking University China School of Electronic Engineering and Computer Science Peking University China National Engineering Laboratory for Big Data Analysis and Application Peking University China

We propose Edit-History Vis, a visual analytics system designed to facilitate interactive exploration on Wikipedia edit history at a fine-grained level. The examination of detailed changes in Wikipedia articles is crucial for understanding how authors’ perspectives vary and conflict during the collaborative editing process. However, it is challenging to reveal the details while preserving the heterogeneous attributes of revisions, namely the time, content, and editor. The Edit-History Vis system integrates editor and textual changes of revisions by utilizing a force-directed revision graph that groups revisions based on standpoints. Through this revision graph, users can identify and analyze editing events such as edit wars, vandalism, repair, and normal updates. The effectiveness of the system in analyzing the edit history is validated through a qualitative comparison with prior work and a quantitative rating from a user study.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Supervised and Unsupervised Coupled Generative Adversarial Networks in Multivariate Time Series

Supervised and Unsupervised Coupled Generative Adversarial N...

引用

Information Science, Parallel and Distributed Systems (ISPDS), International Conference on

作者： Qinfen Wang Yifei Li Yong Xia National Engineering Laboratory for Integrated Aero-Space-Ground-Ocean Big Data Application Technology School of Computer Science and Engineering Northwestern Polytechnical University Xi'an China

Generative adversarial networks (GAN) have been successful in demonstrating the efficiency of generative time series. In this paper, we proposed a novel end-to-end supervised and unsupervised coupled generative adversarial networks (SUC-GAN) framework that consists of a gated recurrent unit (GRU) for generating multivariate time series. We test SUC-GAN on three datasets (i.e. Sines, Energy, Stocks) to investigate the efficacy and properties. The preliminary results show the potential power of supervised, unsupervised, and coupled learning for data generation.

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：