检索结果-内蒙古大学图书馆

Personal Augmented Reality for Information visualization on large Interactive Displays

ieee TRANSACTIONS ON visualization AND COMPUTER GRAPHICS 2021年第2期27卷 1182-1192页

作者： Reipschlager, Patrick Flemisch, Tamara Dachselt, Raimund Tech Univ Dresden Interact Media Lab Dresden Germany Tech Univ Dresden Ctr Tactile Internet CeTi Dresden Germany Tech Univ Dresden Cluster Excellence Phys Lile Dresden Germany

In this work we propose the combination of large interactive displays with personal head-mounted Augmented Reality (AR) for information visualization to facilitate data exploration and analysis. Even though large displays provide more display space, they are challenging with regard to perception, effective multi-user support, and managing data density and complexity. To address these issues and illustrate our proposed setup, we contribute an extensive design space comprising first, the spatial alignment of display, visualizations, and objects in AR space. Next, we discuss which parts of a visualization can be augmented. Finally, we analyze how AR can be used to display personal views in order to show additional information and to minimize the mutual disturbance of data analysts. Based on this conceptual foundation, we present a number of exemplary techniques for extending visualizations with AR and discuss their relation to our design space. We further describe how these techniques address typical visualization problems that we have identified during our literature research. To examine our concepts, we introduce a generic AR visualization framework as well as a prototype implementing several example techniques. In order to demonstrate their potential, we further present a use case walkthrough in which we analyze a movie data set. From these experiences, we conclude that the contributed techniques can be useful in exploring and understanding multivariate data. We are convinced that the extension of large displays with AR for information visualization has a great potential for data analysis and sense-making.

关键词： data visualization data analysis Augmented reality Three-dimensional displays Navigation visualization Augmented Reality Information visualization InfoVis large Displays Immersive Analytics Physical Navigation Multiple Coordinated Views

来源：评论

学校读者我要写书评

暂无评论

LAND USE CLASSIFICATION EFFICIENT VISION TRANSFORMER

LAND USE CLASSIFICATION EFFICIENT VISION TRANSFORMER

引用

ieee International Geoscience and Remote Sensing symposium (IGARSS)

作者： Depoian, Arthur C., II Bailey, Colleen P. Guturu, Parthasarathy Univ North Texas Denton TX 76205 USA

ISBN: (纸本)9798350320107

Land use classification has been a topic of great interest for many years. The ability to automatically recognize land cover as well as detect changes in remote sensing images is important for a variety of industries. Due in part to the size of satellite images, as well as the inefficiency of current methods, machine learning solutions are often very large. With a growing need for on device analysis, it is becoming more important to perform accurate processing without over utilizing resources. To meet this growing need, this paper presents an efficient algorithm for land use classification. The efficacy of the proposed solution is verified through studies to accurately classify remotely sensed images within the EuroSAT dataset.

关键词： Efficient AI Vision Transformer data Augmentation Land Use Classification

来源：评论

学校读者我要写书评

暂无评论

Aerial Images Meet Crowdsourced Trajectories: A New Approach to Robust Road Extraction

引用

ieee TRANSACTIONS ON NEURAL NETWORKS AND LEARNING SYSTEMS 2023年第7期34卷 3308-3322页

作者： Liu, Lingbo Yang, Zewei Li, Guanbin Wang, Kuo Chen, Tianshui Lin, Liang Sun Yat Sen Univ Sch Comp Sci & Engn Guangzhou 510000 Peoples R China Sun Yat Sen Univ Sch Math Guangzhou 510000 Peoples R China Guangdong Univ Technol Sch Informat Engn Guangzhou 510006 Peoples R China

Land remote-sensing analysis is a crucial research in earth science. In this work, we focus on a challenging task of land analysis, i.e., automatic extraction of traffic roads from remote-sensing data, which has widespread applications in urban development and expansion estimation. Nevertheless, conventional methods either only utilized the limited information of aerial images, or simply fused multimodal information (e.g., vehicle trajectories), thus cannot well recognize unconstrained roads. To facilitate this problem, we introduce a novel neural network framework termed cross-modal message propagation network (CMMPNet), which fully benefits the complementary different modal data (i.e., aerial images and crowdsourced trajectories). Specifically, CMMPNet is composed of two deep autoencoders for modality-specific representation learning and a tailor-designed dual enhancement module for cross-modal representation refinement. In particular, the complementary information of each modality is comprehensively extracted and dynamically propagated to enhance the representation of another modality. Extensive experiments on three real-world benchmarks demonstrate the effectiveness of our CMMPNet for robust road extraction benefiting from blending different modal data, either using image and trajectory data or image and light detection and ranging (LiDAR) data. From the experimental results, we observe that the proposed approach outperforms current state-of-the-art methods by large margins. Our source code is resealed on the project page http://***/multimodal_road_***.

关键词： Roads Trajectory Feature extraction data mining Remote sensing Geoscience visualization Aerial images crowdsourced trajectories land remote sensing road network extraction

来源：评论

学校读者我要写书评

暂无评论

Tiny Object Tracking: A large-Scale dataset and a Baseline

引用

ieee TRANSACTIONS ON NEURAL NETWORKS AND LEARNING SYSTEMS 2024年第8期35卷 10273-10287页

作者： Zhu, Yabin Li, Chenglong Liu, Yao Wang, Xiao Tang, Jin Luo, Bin Huang, Zhixiang Anhui Univ Key Lab Electromagnet Environm Sensing Anhui Highe Sch Elect & Informat Engn Key Lab Intelligent Comp & Signal ProcessingMinis Hefei 230601 Peoples R China Anhui Univ Sch Artificial Intelligence Anhui Prov Key Lab Multimodal Cognit Computat Informat Mat & Intelligent Sensing Lab Anhui Prov Hefei 230601 Peoples R China Anhui Univ Sch Comp Sci & Technol Anhui Prov Key Lab Multimodal Cognit Computat Informat Mat & Intelligent Sensing Anhui Prov Hefei 230601 Peoples R China

Tiny objects, frequently appearing in practical applications, have weak appearance and features, and receive increasing interests in many vision tasks, such as object detection and segmentation. To promote the research and development of tiny object tracking, we create a large-scale video dataset, which contains 434 sequences with a total of more than 217K frames. Each frame is carefully annotated with a high-quality bounding box. In data creation, we take 12 challenge attributes into account to cover a broad range of viewpoints and scene complexities, and annotate these attributes for facilitating the attribute-based performance analysis. To provide a strong baseline in tiny object tracking, we propose a novel multilevel knowledge distillation network (MKDNet), which pursues three-level knowledge distillations in a unified framework to effectively enhance the feature representation, discrimination, and localization abilities in tracking tiny objects. Extensive experiments are performed on the proposed dataset, and the results prove the superiority and effectiveness of MKDNet compared with state-of-the-art methods. The dataset, the algorithm code, and the evaluation code are available at https://***/mmic-lcl/datasets-and-benchmark-code.

关键词： Object tracking visualization Annotations Knowledge engineering Location awareness Target tracking Sensors Benchmark dataset knowledge distillation tiny objects visual tracking

来源：评论

学校读者我要写书评

暂无评论

EFFICIENT large-SCALE DAMAGE ASSESSMENT AFTER NATURAL DISASTERS WITH UAVS AND DEEP LEARNING

EFFICIENT LARGE-SCALE DAMAGE ASSESSMENT AFTER NATURAL DISAST...

引用

ieee International Geoscience and Remote Sensing symposium (IGARSS)

作者： Rahnemoonfar, Maryam Safavi, Farshad Lehigh Univ Dept Comp Sci & Engn Bethlehem PA 18015 USA Lehigh Univ Dept Civil & Environm Engn Bethlehem PA 18015 USA Univ Maryland Dept Comp Sci Baltimore MD USA

ISBN: (纸本)9798350320107

Frequent and increasingly severe natural disasters due to climate change threaten human health and infrastructure. The provision of accurate, timely, and understandable information has the potential to revolutionize disaster management. While traditional analyses provide some insights into the data, the complexity, scale, and multi-disciplinary nature of the data necessitate advanced, intelligent solutions. The main barrier is the lack of near real-time data analysis. Recently there has been a surge of practical applications of real-time semantic segmentation. Real-time semantic segmentation requires fast and high-quality predictions. As a result, lightweight architectures with low latency and computational costs are necessary for efficient semantic segmentation. In this article, we developed several encoder-decoder and two-pathway architectures and compared their performance on a novel RescuNet dataset.

关键词： Deep learning Real-time semantic segmentation Natural disasters Unmanned Aerial Vehicles

来源：评论

学校读者我要写书评

暂无评论

Preliminary Exploration of Oil-immersed Inverted Current Transformer Partial Discharge Identification by Wide Frequency Device 2

Preliminary Exploration of Oil-immersed Inverted Current Tra...

引用

ieee 2nd International Conference on Power Science and Technology (ICPST)

作者： Qian, Dun Yang, Zhi Feng, Yuzhe State Grid Zhejiang Elect Power Co Ltd Elect Power Res Inst Hangzhou Peoples R China

ISBN: (纸本)9798350349047;9798350349030

Oil-immersed inverted current transformers are widely used in power systems with voltage levels of 110kV and above. The insulation breakdown of oil-immersed inverted current transformer leads to a large explosion hazard. In order to solve this problem, deployed a wide frequency device for partial discharges based on the pulse current method. The article presents a preliminary exploration of partial discharge identification in oil-immersed inverted current transformers using a wide frequency device. The device employs the pulse current method to capture partial discharge data, which is then analyzed through cluster analysis, algorithmic judgment, and data visualization with manual judgment. The approach offers a promising tool for enhancing the safety and reliability of these critical electrical equipment.

关键词： Oil-immersed inverted current transformer Current transformer Partial Discharge Phase Resolved Partial Discharge(PRPD) mapping cluster analysis PRPD historical mapping

来源：评论

学校读者我要写书评

暂无评论

Self-Supervised Learning for Heterogeneous Audiovisual Scene analysis

引用

ieee TRANSACTIONS ON MULTIMEDIA 2023年 25卷 3534-3545页

作者： Hu, Di Wang, Zheng Nie, Feiping Wang, Rong Li, Xuelong Renmin Univ China Beijing Key Lab Big Data Management & Anal Method Gaoling Sch Artificial Intelligence Beijing 100872 Peoples R China Xi An Jiao Tong Univ Sch Software Engn Xian 10698 Peoples R China Northwestern Polytech Univ Sch Artificial Intelligence Opt & Elect iOPEN Xian 710072 Shaanxi Peoples R China Northwestern Polytech Univ Key Lab Intelligent Interact & Applicat Minist Ind & Informat Technol Xian 710072 Shaanxi Peoples R China

Due to the difficulty of annotating large amounts of training data, directly learning the association of sound and its makers in natural videos is a challenging task for machines. In this paper, we present a novel audiovisual model that introduces a soft-clustering module as the audio and visual content detector, and regards the pervasive property of audiovisual concurrency as the latent supervision for inferring the correlation among detected contents. Furthermore, we discover for the first time that the complexity of data has an impact on the training efficiency and subsequent performance of audiovisual model, i.e., more complex data brings more obstacles to the model training, and degrades the performance of downstream audiovisual tasks. To address the issue of audiovisual learning, we propose a novel heterogeneous audiovisual scene analysis module that trains the model from simple to complex scene. We show that such ordered learning procedure rewards the model the merits of easy training and fast convergence. Meanwhile, our audiovisual model can also provide effective unimodal representation and cross-modal alignment performance. We further deploy the well-trained model into practical audiovisual sound localization and separation tasks. We show that our localization model significantly outperforms existing methods, based on which we show comparable performance in sound separation task by comparison to several related SOTA audiovisual learning methods without referring external visualsupervision.

关键词： visualization Task analysis Location awareness Complexity theory Training Concurrent computing Image analysis Audiovisual sound localization and separation heterogenous audiovisual scene analysis multimodal audiovisual learning

来源：评论

学校读者我要写书评

暂无评论

MONITORING THE THAW SLUMP-DERIVED THERMOKARST BY ALOS-2 INTERFEROMETRIC data IN PERMAFROST TERRAIN OF QINGHAI-TIBET PLATEAU BETWEEN 2015 AND 2022

MONITORING THE THAW SLUMP-DERIVED THERMOKARST BY ALOS-2 INTE...

引用

ieee International Geoscience and Remote Sensing symposium (IGARSS)

作者： Zou, Lichuan Wang, Chao Zhang, Bo Zhang, Zhengjia Tang, Yixian Zhang, Hong Chinese Acad Sci Aerosp Informat Res Inst Key Lab Digital Earth Sci Beijing 100094 Peoples R China Int Res Ctr Big Data Sustainable Dev Goals Beijing 100094 Peoples R China Univ Chinese Acad Sci Beijing 100049 Peoples R China China Univ Geosci Sch Geog & Informat Engn Wuhan 430074 Peoples R China

ISBN: (纸本)9798350320107

With the global warming, thaw slump activity has increased in permafrost regions of Qinghai-Tibet Plateau (QTP), which influence the stability of human infrastructure and carbon cycling. However, the intrinsic dynamic process of surface displacement of the retrogression thaw slump (RTS) is still less understood. Here, we employed the spaceborne interferometric synthetic aperture radar (InSAR) based on L-band ALOS-2 PLASAR2 data acquired from December 2015 to May 2022 to monitor the surface subsidence trends of thaw slump-derived thermokarst in permafrost terrain of Qinghai-Tibet Plateau (QTP). The InSAR analysis reveals the thermokarst subsidence of -81 similar to 46 mm/year between 2015 and 2022. A large number of RTS were distributed in the slopes, with the large annual average sedimentation rates. Besides, the time-series seasonal deformation reveals that during the period from January 2019 to March 2019, with the mean temperature below 0 degrees C, RTS' deformation still represents large seasonal subsidence, which indicates the intrinsic pattern of surface displacement of the thaw slump is not simple cold-season freeze heaving.

关键词： Thaw slump Permafrost InSAR ALOS-2 Tibetan Plateau

来源：评论

学校读者我要写书评

暂无评论

EventsVista: Enhancing Event visualization and Interpretation 10

EventsVista: Enhancing Event Visualization and Interpretatio...

引用

10th ieee International Conference on Behavioural and Social Computing, BESC 2023

作者： Farhat, Mohamad Khalil Zhang, Ji Tao, Xiaohui Li, Tianning Yu, Ting University of Southern Queensland Australia Zhejiang Lab China

ISBN: (纸本)9798350395884

The surge in technological advancements in the last decade has given rise to massive amounts of data. Researchers have since developed a wide range of software to visualize extensive data sets. data visualization techniques ensure an enhanced analysis of substantial amounts of information and bypass the difficulty in understanding the data they reflect. These techniques facilitate the collating of information and detecting the overall trends of events, which plays a critical role in different fields such as finance, medical recording, cybersecurity, and essential day-To-day tasks. However, information can be poorly translated when scaling from large datasets to representative graphics, either by causing information loss or over-crowding the audience receiving the visual in any application or software. Understanding and using the appropriate methodologies becomes critical for the authentic translation of data as well as the engagement of the audience. In this paper, we demonstrate the evolution of big data, the key elements of data visualization, and the operations conventionally used in visualizing the space-Time cube. Moreover, we introduce our newly developed data visualization software, EventsVista, which displays and analyzes events based on the assessment of previous data visualization methods. EventsVista explicitly demonstrates the effect of merging conventional operations together, as well as adding novel ones based on the Degree of Freedom (DoF) strategy. Furthermore, EventsVista was tested on events taking place at the University of Southern Queensland, Toowoomba, Australia, prior to its launch for public user assessment. Positive user experience reflected the success of the DoF strategy in improving the event's interpretation using our distinctive and publicly accessible visualization application. © 2023 ieee.

关键词： data visualization

来源：评论

学校读者我要写书评

暂无评论

Towards LLM-Enhanced Digital Twins of Intelligent Computing Center 24

Towards LLM-Enhanced Digital Twins of Intelligent Computing ...

引用

24th ieee International Conference on Software Quality, Reliability and Security Companion, QRS-C 2024

作者： Yang, Yi Liu, Manxian Jia, Kun Dong, Wenjie Wang, Jian Institute of Automation Chinese Academy of Sciences Beijing China The 33rd Research Institute of China Electronics Technology Group Corporation Taiyuan China

ISBN: (纸本)9798350365658

Digital twin techniques enable the digital modeling of the power and environmental systems of intelligent computing systems. However, the traditional analysis methods of digital twins do not integrated expert experience with data-driven approaches flexibly. They also analyze the information from multiple dimensions in isolation without considering their relations. This paper proposes a large language model-enhanced digital twin system for addressing these issues. The proposed system provides natural language-based interactive analysis, allowing experts to collaborate with algorithms for risk analysis;it employs a large language model-driven multi-agent system to simulate the interactions of the dynamic environmental systems, achieving multi-dimensional joint prediction. Additionally, this paper presents a case study to demonstrate the usefulness and effectiveness of the proposed system. © 2024 ieee.

关键词： Risk analysis

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：