检索结果-内蒙古大学图书馆

arXiv 2023年

作者： Zhang, Xinyu Wang, Li Chen, Jian Fang, Cheng Yang, Lei Song, Ziying Yang, Guangqi Wang, Yichen Zhang, Xiaofei Li, Jun Li, Zhiwei Yang, Qingshan Zhang, Zhenlin Ge, Shuzhi Sam State Key Laboratory of Automotive Safety and Energy the School of Vehicle and Mobility Ts-inghua University Beijing100084 China School of Mechanical and Electrical Engineering China University of Mining and Technology-Beijing Beijing100083 China School of Artificial Intelligence China University of Mining and Technology-Beijing Beijing100083 China College of Information Science and Technology Beijing University of Chemical Technology Beijing100029 China Beijing Jingwei Hirain Technologies Co. Inc. Beijing100191 China China Automotive Innovation Cooperation Nanjing211113 China Department of Electrical and Computer Engineering Interactive Digital Media Institute Social Robotics Laboratory National University of Singapore Singapore

Radar has stronger adaptability in adverse scenarios for autonomous driving environmental perception compared to widely adopted cameras and LiDARs. Compared with commonly used 3D radars, latest 4D radars have precise vertical resolution and higher point cloud density, making it a highly promising sensor for autonomous driving in complex environmental perception. However, due to the much higher noise than LiDAR, manufacturers choose different filtering strategies, resulting in a direct ratio between point cloud density and noise level. There is still a lack of comparative analysis on which method is beneficial for deep learning-based perception algorithms in autonomous driving. One of the main reasons is that current datasets only adopt one type of 4D radar, making it difficult to compare different 4D radars in the same scene. Therefore, in this paper, we introduce a novel large-scale multi-modal dataset featuring, for the first time, two types of 4D radars captured simultaneously. This dataset enables further research into effective 4D radar perception algorithms. Our dataset consists of 151 consecutive series, most of which last 20 seconds and contain 10,007 meticulously synchronized and annotated frames. Moreover, our dataset captures a variety of challenging driving scenarios, including many road conditions, weather conditions, nighttime and daytime with different lighting intensities and periods. Our dataset annotates consecutive frames, which can be applied to 3D object detection and tracking, and also supports the study of multi-modal tasks. We experimentally validate our dataset, providing valuable results for studying different types of 4D radars. This dataset is released on https://***/adeptthu/Dual-Radar. Copyright © 2023, The Authors. All rights reserved.

关键词： Autonomous vehicles

来源：评论

学校读者我要写书评

暂无评论

Digital Map Generation Workflow Demonstrated on ZalaZONE Automotive Proving Ground Elements

Digital Map Generation Workflow Demonstrated on ZalaZONE Aut...

引用

IEEE Symposium on Network Operations and Management

作者： Árpád Somogyi Tamás Tettamanti Pál Varga Zsolt Szalay Dániel Baranyai Tamás Lovas Dept. of Photogrammetry and Geoinformatics Fac. of Civil Engineering Budapest University of Technology and Economics Budapest Hungary Systems and Control Laboratory Institute for Computer Science and Control of the Eötvös Loránd Research Network Budapest Hungary Dept. of Telecommunications and Media Informatics Fac. of Electrical Engineering and Informatics Budapest University of Technology and Economics Budapest Hungary Dept. of Automotive Technologies Fac. of Transportation Eng. and Vehicle Eng. Budapest University of Technology and Economics Budapest Hungary

One major goal of digital twin technology applied in the Architecture, engineering, and Construction (AEC) Industry is the mapping of roads and road environments with their associated information. Such digital twins can support cost-effective testing, simulations, and the production of mapping backgrounds for automated vehicles and generally for Cooperative, Connected and Automated Mobility (CCAM) developments. The efficient workflow that leads from data acquisition to finished products depends on a number of requirements and knowledge sets. The main contribution of the paper is a general methodology of creating HD-map from point cloud and image data set. Accordingly, the most widely used survey techniques are presented and the requirements for the objects to be derived are summarized. The paper then outlines a general implementation procedure as an expert knowledge through a real-life example, where the results obtained for the ZalaZONE Automotive Proving Ground are demonstrated.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Relationship between number of annotations and accuracy in segmentation of the erector spinae muscle using Bayesian U-Net in torso CT images

Relationship between number of annotations and accuracy in s...

引用

International Forum on Medical Imaging in Asia 2021, IFMIA 2021

作者： Wakamatsu, Yuichi Kamiya, Naoki Zhou, Xiangrong Hara, Takeshi Fujita, Hiroshi Department of Media Informatics Graduate School of Information Science and Technology Aichi Prefectural University 1522-3 Ibaragabasama Nagakute-shi Aichi480-1198 Japan Department of Information Systems Graduate School of Information Science and Technology Aichi Prefectural University 1522-3 Ibaragabasama Nagakute-shi Aichi480-1198 Japan Department of Electrical Electronic and Computer Engineering Faculty of Engineering Gifu University 1-1 Yanagido Gifu-shi Gifu501-1193 Japan

ISBN: (数字)9781510644212

ISBN: (纸本)9781510644205

Supervised learning for image segmentation requires annotated images. However, image annotation has the problem that it is time-consuming. This problem is particularly significant in the erector spinae muscle segmentation due to the large size of the muscle. Therefore, this study considers the relationship between the number of annotated images used for training and segmentation accuracy of the erector spinae muscle in torso CT images. We use Bayesian U-Net, which has shown high accuracy in thigh muscle segmentation, for the segmentation of the erector spinae muscle. In the network training, we limit the number of slices for each case and the number of cases to 100%, 50%, 25%, and 10%. In the experiment, we use 30 torso CT images, including 6 cases for the test dataset. Experimental results are evaluated by the mean Dice value of the test dataset. Using 100% of the slices per case, the segmentation accuracy with 100%, 50%, 25%, and 10% of the cases was 0.934, 0.927, 0.926, and 0.890, respectively. On the other hand, using 100% of the cases, the segmentation accuracy with 100%, 50%, 25%, and 10% of the slices per case was 0.934, 0.934, 0.933, and 0.931, respectively. Furthermore, the segmentation accuracy with 100% of the cases and 10% of the slices per case was higher than that of the previous method. We showed that it is feasible to achieve high segmentation accuracy with a limited number of annotated images by selecting several slices from a limited number of cases for training. © 2021 SPIE.

关键词： Image segmentation

来源：评论

学校读者我要写书评

暂无评论

Shaping Integrity: Why Generative Artificial Intelligence Does Not Have to Undermine Education

arXiv

引用

arXiv 2024年

作者： Tan, Myles Joshua Toledo Maravilla, Nicholle Mae Amor Tan Department of Electrical and Computer Engineering Herbert Wertheim College of Engineering University of Florida GainesvilleFL32611 United States Department of Epidemiology College of Public Health & Health Professions College of Medicine University of Florida GainesvilleFL32610 United States Biology Program College of Arts and Sciences University of St. La Salle Negros Occidental Bacolod City6100 Philippines Department of Natural Sciences College of Arts and Sciences University of St. La Salle Negros Occidental Bacolod City6100 Philippines Department of Chemical Engineering College of Engineering and Technology University of St. La Salle Negros Occidental Bacolod City6100 Philippines Department of Electronics Engineering College of Engineering and Technology University of St. La Salle Negros Occidental Bacolod City6100 Philippines Yo-Vivo Corporation Negros Occidental Bacolod City6100 Philippines

This paper examines the role of generative artificial intelligence (GAI) inpromoting academic integrity within educational settings. It explores how AIcan be ethically integrated into classrooms to enhance learning experiences,foster intrinsic motivation, and support voluntary behavior change amongstudents. By analyzing established ethical frameworks and educational theoriessuch as deontological ethics, consequentialism, constructivist learning, andSelf-Determination Theory (SDT), the paper argues that GAI, when usedresponsibly, can enhance digital literacy, encourage genuine knowledgeconstruction, and uphold ethical standards in education. This researchhighlights the potential of GAI to create enriching, personalized learningenvironments that prepare students to navigate the complexities of the modernworld ethically and effectively. © 2024, CC BY-NC-SA.

关键词： Contrastive Learning

来源：评论

学校读者我要写书评

暂无评论

Machine learning traction force maps of cell monolayers

arXiv

引用

arXiv 2023年

作者： Li, Changhao Feng, Luyi Park, Yang Jeong Yang, Jian Li, Ju Zhang, Sulin Department of Engineering Science and Mechanics Pennsylvania State University University ParkPA United States Department of Nuclear Science and Engineering Department of Materials Science and Engineering Massachusetts Institute of Technology CambridgeMA02139 United States Department of Electrical and Computer Engineering Institute of New Media and Communications Seoul National University Gwanak-gu Seoul08826 Korea Republic of Department of Biomedical Engineering Pennsylvania State University University ParkPA United States Department of Material Science and Engineering Pennsylvania State University University ParkPA United States

Cellular force transmission across a hierarchy of molecular switchers is central to mechanobiological responses. However, current cellular force microscopies suffer from low throughput and resolution. Here we introduce and train a generative adversarial network (GAN) to paint out traction force maps of cell monolayers with high fidelity to the experimental traction force microscopy (TFM). The GAN analyzes traction force maps as an image-to-image translation problem, where its generative and discriminative neural networks are simultaneously cross-trained by hybrid experimental and numerical datasets. In addition to capturing the colony-size and substrate-stiffness dependent traction force maps, the trained GAN predicts asymmetric traction force patterns for multicellular monolayers seeding on substrates with stiffness gradient, implicating collective durotaxis. Further, the neural network can extract experimentally inaccessible, the hidden relationship between substrate stiffness and cell contractility, which underlies cellular mechanotransduction. Trained solely on datasets for epithelial cells, the GAN can be extrapolated to other contractile cell types using only a single scaling factor. The digital TFM serves as a high-throughput tool for mapping out cellular forces of cell monolayers and paves the way toward data-driven discoveries in cell mechanobiology. © 2023, CC BY.

关键词： Generative adversarial networks

来源：评论

学校读者我要写书评

暂无评论

Flexible dual-branched message passing neural network for quantum mechanical property prediction with molecular conformation

arXiv

引用

arXiv 2021年

作者： Jo, Jeonghee Kwak, Bumju Lee, Byunghan Yoon, Sungroh Interdisciplinary Program in Bioinformatics Seoul National University Recommendation Team Kakao Corporation Department of Electronic and IT Media Engineering Seoul National University of Science and Technology Department of Electrical and Computer Engineering Seoul National University

A molecule is a complex of heterogeneous components, and the spatial arrangements of these components determine the whole molecular properties and characteristics. With the advent of deep learning in computational chemistry, several studies have focused on how to predict molecular properties based on molecular configurations. Message passing neural network provides an effective framework for capturing molecular geometric features with the perspective of a molecule as a graph. However, most of these studies assumed that all heterogeneous molecular features, such as atomic charge, bond length, or other geometric features always contribute equivalently to the target prediction, regardless of the task type. In this study, we propose a dual-branched neural network for molecular property prediction based on message-passing framework. Our model learns heterogeneous molecular features with different scales, which are trained flexibly according to each prediction target. In addition, we introduce a discrete branch to learn single atom features without local aggregation, apart from message-passing steps. We verify that this novel structure can improve the model performance with faster convergence in most targets. The proposed model outperforms other recent models with sparser representations. Our experimental results indicate that in the chemical property prediction tasks, the diverse chemical nature of targets should be carefully considered for both model performance and generalizability. © 2021, CC BY.

关键词： Deep learning

来源：评论

学校读者我要写书评

暂无评论

Efficient Computing of Disaster-Disjoint Paths: Greedy and Beyond

Efficient Computing of Disaster-Disjoint Paths: Greedy and B...

引用

IEEE Conference on computer Communications Workshops, INFOCOM Wksps

作者： Balázs Vass Erika Bérczi-Kovács Péter Gyimesi János Tapolcai Department of Telecommunication and Media Informatics Faculty of Electrical Engineering and Informatics (VIK) Budapest University of Technology and Economics (BME) and HUN-REN-BME Information Systems Research Group and MTA-BME Future Internet Research Group Faculty of Mathematics and Computer Science Babeş-Bolyai University Cluj Napoca Romania ELTE Eötvös Loránd University Budapest Hungary HUN-REN-ELTE Egerváry Research Group on Combinatorial Optimization

ISBN: (数字)9798350384475

ISBN: (纸本)9798350384482

In a network topology $G$ , we say a set of st-paths are disaster-disjoint if no disaster strikes more than one path. In this poster, we explore the basic capabilities and limitations of greedy and more advanced algorithms for computing maximal collections of such paths in planar networks. An algorithm is greedy if it generates consecutive paths $P_{1}, P_{2},\ldots$ according to a simple rule. In the simplest setting, the only rule is that $P_{i+1}$ is the closest clockwise disaster-disjoint from $P_{i}$ . We find that the simplest greedy may fail even when 1) $G$ is planar, 2) each disaster region is connected, and 3) each node failure (apart from s and $t$ ) is considered possible. Adding a simple rule explained in [1] yields a correct polynomial-time algorithm for the above problem. Finally, we digest a recent related near-linear runtime algorithm of [2] solving a more general problem and discuss the underlying relations among the foundations of these algorithms.

关键词： Greedy algorithms Visualization Runtime Network topology Disasters Conferences

来源：评论

学校读者我要写书评

暂无评论

Enhancing 3D Object Detection in Autonomous Vehicles Based on Synthetic Virtual Environment Analysis

arXiv

引用

arXiv 2024年

作者： Li, Vladislav Siniosoglou, Ilias Karamitsou, Thomai Lytos, Anastasios Moscholios, Ioannis D. Goudos, Sotirios K. Banerjee, Jyoti S. Sarigiannidis, Panagiotis Argyriou, Vasileios Kingston University Department of Networks and Digital Media Kingston upon Thames United Kingdom University of Western Macedonia Department of Electrical and Computer Engineering Kozani Greece MetaMind Innovations P.C. R"&D Department Kozani Greece Sidroco Holdings Ltd. Nicosia Cyprus University of Peloponnese Department of Informatics and Telecommunications Tripoli Greece Aristotle University of Thessaloniki Physics Department Thessaloniki Greece Bengal Institute of Technology Kolkata India

Autonomous Vehicles (AVs) use natural images and videos as input to understand the real world by overlaying and inferring digital elements, facilitating proactive detection in an effort to assure safety. A crucial aspect of this process is real-time, accurate object recognition through automatic scene analysis. While traditional methods primarily concentrate on 2D object detection, exploring 3D object detection, which involves projecting 3D bounding boxes into the three-dimensional environment, holds significance and can be notably enhanced using the AR ecosystem. This study examines an AI model’s ability to deduce 3D bounding boxes in the context of real-time scene analysis while producing and evaluating the model’s performance and processing time, in the virtual domain, which is then applied to AVs. This work also employs a synthetic dataset that includes artificially generated images mimicking various environmental, lighting, and spatiotemporal states. This evaluation is oriented in handling images featuring objects in diverse weather conditions, captured with varying camera settings. These variations pose more challenging detection and recognition scenarios, which the outcomes of this work can help achieve competitive results under most of the tested conditions. © 2024, CC BY-NC-ND.

关键词： Virtual environments

来源：评论

学校读者我要写书评

暂无评论

GATE: Graph CCA for Temporal SElf-supervised Learning for Label-efficient fMRI Analysis

arXiv

引用

arXiv 2022年

作者： Peng, Liang Wang, Nan Xu, Jie Zhu, Xiaofeng Li, Xiaoxiao Center for Future Media Department of Computer Science and Engineering University of Electronic Science and Technology of China Chengdu610054 China School of Computer Science and Engineering East China Normal University Shanghai200062 China The University of British Columbia VancouverBCV6T 1Z4 Canada School of Computer Science and Engineering University of Electronic Science and Technology of China Chengdu610056 China The Shenzhen Institute for Advanced Study University of Electronic Science and Technology of China Shenzhen China Electrical and Computer Engineering The University of British Columbia VancouverBCV6T 1Z4 Canada

In this work, we focus on the challenging task, neuro-disease classification, using functional magnetic resonance imaging (fMRI). In population graph-based disease analysis, graph convolutional neural networks (GCNs) have achieved remarkable success. However, these achievements are inseparable from abundant labeled data and sensitive to spurious signals. To improve fMRI representation learning and classification under a label-efficient setting, we propose a novel and theory-driven self-supervised learning (SSL) framework on GCNs, namely Graph CCA for Temporal sElf-supervised learning on fMRI analysis (GATE). Concretely, it is demanding to design a suitable and effective SSL strategy to extract formation and robust features for fMRI. To this end, we investigate several new graph augmentation strategies from fMRI dynamic functional connectives (FC) for SSL training. Further, we leverage canonical-correlation analysis (CCA) on different temporal embeddings and present the theoretical implications. Consequently, this yields a novel two-step GCN learning procedure comprised of (i) SSL on an unlabeled fMRI population graph and (ii) fine-tuning on a small labeled fMRI dataset for a classification task. Our method is tested on two independent fMRI datasets, demonstrating superior performance on autism and dementia diagnosis. Our code is available at https://***/LarryUESTC/GATE. Copyright © 2022, The Authors. All rights reserved.

关键词： Graphic methods

来源：评论

学校读者我要写书评

暂无评论

Energy-efficient Beamforming Design with Partial CSI Feedback for RIS-assisted Systems

引用

IEEE Transactions on Wireless Communications 2025年

作者： Ge, Xiaochun Shen, Wenqian Shim, Byonghyo Guan, Yong Liang An, Jianping Beijing Institute of Technology School of Information and Electronics Beijing100081 China Seoul National University Institute of New Media and Communications Department of Electrical and Computer Engineering Seoul08826 Korea Republic of Nanyang Technological University School of Electrical and Electronics Engineering 639798 Singapore

This paper investigates beamforming design with partial channel state information (CSI) feedback aimed at maximizing the energy efficiency (EE) of a reconfigurable intelligent surface (RIS)-assisted millimeter-wave (mmWave) multi-user system. By leveraging the spatial reciprocity, which means that the path angle information (PAI) and power angular spectrum (PAS) of sparse paths are similar in the uplink and downlink channels, we can acquire the downlink PAI and PAS at the base station (BS) via uplink estimation. Subsequently, we select several paths that contribute to EE maximization from all the cascaded paths and define them as dominant paths. Consequently, only the small-scale fading coefficients (i.e., the normalized path gain information, NPGI) of these selected dominant paths need to be fed back from the user equipments (UEs), thereby significantly reducing the feedback overhead. Moreover, we update the active BS beamformer and passive RIS beamformer by using the feedback of partial NPGI to further improve the EE. Numerical results demonstrate the superiority of our proposed algorithms over conventional schemes. © 2002-2012 IEEE.

关键词： Beamforming

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：