检索结果-内蒙古大学图书馆

43rd Annual Conference on European Association for computer graphics, EUROgraphics 2022

作者： Müller-Huschke, J. Ritter, M. Harders, M. Interactive Graphics and Simulation Group The Department of Computer Science University of Innsbruck Austria

ISBN: (纸本)9783038681694

Most current research on automatically captioning and describing scenes with spatial content focuses on images. We outline that generating descriptive text for a synthesized 3D scene can be achieved via a suitable intermediate representation employed in the synthesis algorithm. As an example, we synthesize scenes of medieval village settings, and generate their descriptions. Our system employs graph grammars, Markov Chain Monte Carlo optimization, and a natural language generation pipeline. Randomly placed objects are evaluated and optimized by a cost function capturing neighborhood relations, path layouts, and collisions. Further, in a pilot study we assess the performance of our framework by comparing the generated descriptions to others provided by human subjects. While the latter were often short and low-effort, the highest-rated ones clearly outperform our generated ones. Nevertheless, the average of all collected human descriptions was indeed rated by the study participants as being less accurate than the automated ones. © 2022 The Author(s).

关键词： Cost functions

来源：评论

学校读者我要写书评

暂无评论

Stereo Conversion with Disparity-Aware Warping, Compositing and Inpainting

Stereo Conversion with Disparity-Aware Warping, Compositing ...

引用

IEEE Workshop on Applications of computer Vision (WACV)

作者： Lukas Mehl Andrés Bruhn Markus Gross Christopher Schroers Institute for Visualization and Interactive Systems University of Stuttgart Department of Computer Science Computer Graphics Lab ETH Zurich DisneyResearch|Studios Switzerland

Despite of exciting advances in image-based rendering and novel view synthesis, it is still challenging to achieve high-resolution results that can reach production-level quality when applying such methods to the task of stereo conversion. At the same time, only very few dedicated stereo conversion approaches exist, which also fall short in terms of the required quality. Hence, in this paper, we present a novel method for high-resolution 2D-to-3D conversion. It is fully differentiable in all of its stages and performs disparity-informed warping, consistent foreground-background compositing, and background-aware inpainting. To enable temporal consistency in the resulting video, we propose a strategy to integrate information from additional video frames. Extensive ablation studies validate our design choices, leading to a fully automatic model that outperforms existing approaches by a large margin (49-70% LPIPS error reduction). Finally, inspired from current practices in manual stereo conversion, we introduce optional interactive tools into our model, which allow to steer the conversion process and make it significantly more applicable for 3D film production.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Enhancing Cyclone center Identification in Radar Images through Deep Learning and Match Recognition 15

Enhancing Cyclone Center Identification in Radar Images thro...

引用

15th International Conference on Computing Communication and Networking Technologies, ICCCNT 2024

作者： Shanmugam, Thirumurugan Chandrashekar, Chirag Subburaj, Maheswari Sivaraman, Arun Kumar Sultana, Ajmery Thangavel, Senthil Kumar University of Technology and Applied Sciences College of Computing and Information Sciences Suhar Oman Vellore Institute of Technology School of Computer Science and Engineering Chennai India Photon Interactive Inc Digital Engineering Solution Center ON Canada Algoma University Brampton School of Computer Science and Technology ON Canada Amrita Vishwa Vidyapeetham Department of Computer Science and Engineering Coimbatore India

ISBN: (纸本)9798350370249

This research focuses on improving the identification of cyclone centers using deep learning and match recognition applied to radar images. Accurately pinpointing the cyclone's center is vital for predicting its intensity and trajectory. However, challenges persist in automatically locating the center due to the diverse nature of cyclone morphology and structure. To address this, the deep convolutional network's capability is leveraged to capture various structural features in images by proposing two-step approach for cyclone center localization. Initially, a pre-trained EfficientDet model is employed using transfer learning to obtain weights. Subsequently, these weights, along with the data, are utilized in a deep learning model to provide precise coordinates of the cyclone center in the respective image. The effectiveness of existing deep learning and machine learning models show that the cyclone prediction systems have an accuracy ranging from 86% to 92% or better, and cyclone eye detection accuracy surpassing 87%. Experiment outcomes indicate that the proposed methodology outperforms conventional methods and existing works, showcasing its potential for enhancing cyclone monitoring and forecasting. © 2024 IEEE.

关键词： Cyclone detection deep learning efficient det and YOLO model transfer learning

来源：评论

学校读者我要写书评

暂无评论

Robust reconstruction of curved line structures in noisy point clouds

引用

Visual Informatics 2021年第3期5卷 1-14页

作者： Marcel Ritter Daniel Schiffner Matthias Harders Interactive Graphics and Simulation Group Department of Computer ScienceUniversity of InnsbruckAustria Airborne Hydromapping GmbH InnsbruckAustria DIPF|Leibniz Institute for Research and Information Education FrankfurtGermany

Point-based geometry representations have become widely used in numerous contexts,ranging from particle-based simulations,over stereo image matching,to depth sensing via light detection and *** application focus is on the reconstruction of curved line structures in noisy 3D point cloud *** algorithms operating on such point clouds often rely on the notion of a local *** the latter,our approach employs multi-scale neighborhoods,for which weighted covariance measures of local points are *** line structures are reconstructed via vector field tracing,using a bidirectional piecewise streamline *** also introduce an automatic selection of optimal starting points via multi-scale geometric *** pipeline development and choice of parameters was driven by an extensive,automated initial analysis process on over a million prototype test *** behavior of our approach is controlled by several parameters—the majority being set automatically,leaving only three to be controlled by a *** an extensive,automated final evaluation,we cover over one hundred thousand parameter sets,including 3D test geometries with varying curvature,sharp corners,intersections,data holes,and systematically applied varying types of ***,we analyzed different choices for the point of reference in the co-variance computation;using a weighted mean performed best in most *** addition,we compared our method to current,publicly available line reconstruction *** to thirty times faster execution times were achieved in some cases,at comparable error ***,we also demonstrate an exemplary application on four real-world 3D light detection and ranging datasets,extracting power line cables.

关键词： Computational geometry Noisy point clouds Line reconstruction Automatic Adaptive control

来源：评论

学校读者我要写书评

暂无评论

Deterministic Ansätze for the Measurement-based Variational Quantum Eigensolver

arXiv

引用

arXiv 2023年

作者： Schroeder, Anna Heller, Matthias Gachechiladze, Mariami Merck KGaA Darmstadt Germany Department of Computer Science Technical University of Darmstadt Darmstadt Germany Fraunhofer Institute for Computer Graphics Research IGD Darmstadt Germany Interactive Graphics Systems Group Technical University of Darmstadt Darmstadt Germany

Measurement-based quantum computing (MBQC) is a promising approach to reducing circuit depth in noisy intermediate-scale quantum algorithms such as the Variational Quantum Eigensolver (VQE). Unlike gate-based computing, MBQC employs local measurements on a preprepared resource state, offering a trade-off between circuit depth and qubit count. Ensuring determinism is crucial to MBQC, particularly in the VQE context, as a lack of flow in measurement patterns leads to evaluating the cost function at irrelevant locations. This study introduces MBVQE-ansätze that respect determinism and resemble the widely used problem-agnostic hardware-efficient VQE ansatz. We evaluate our approach using ideal simulations on the Schwinger Hamiltonian and XY-model and perform experiments on IBM hardware with an adaptive measurement capability. In our use case, we find that ensuring determinism works better via postselection than by adaptive measurements at the expense of increased sampling cost. Additionally, we propose an efficient MBQC-inspired method to prepare the resource state, specifically the cluster state, on hardware with heavy-hex connectivity, requiring a single measurement round, and implement this scheme on quantum computers with 27 and 127 qubits. We observe notable improvements for larger cluster states, although direct gate-based implementation achieves higher fidelity for smaller instances. © 2023, CC BY.

关键词： Qubits

来源：评论

学校读者我要写书评

暂无评论

An Industrial Internet Cloud-Edge Collaborative Scheme and Evaluation System 3

An Industrial Internet Cloud-Edge Collaborative Scheme and E...

引用

3rd International Conference on Cloud Computing, Big Data Application and Software Engineering, CBASE 2024

作者： Sun, Tianwen Wang, Yushen Yang, Guang Xiong, Runqun Yang, Kai Zhang, Wenhua System integration center China Mobile Group Jiangsu Co. Ltd. Nanjing China China Mobile Group Jiangsu Co. Ltd. Group Customer Department Nanjing China School of Computer Science Southeast University Nanjing China Migu Interactive Entertainment Co. Ltd. Cloud network technology Department Nanjing China

ISBN: (纸本)9798331506582

The scale of the global cloud computing market is growing rapidly. In the context of the global digital economy, cloud computing has become an inevitable choice for digital transformation of enterprises. With the rapid development of 5G, Internet of things and other technologies, edge computing has attracted more and more attention, and cloud-edge collaboration has become an important trend. Only cloud computing and edge computing can meet the needs of various scenarios through close collaboration, so as to maximize the application value of cloud computing and edge computing. This paper systematically puts forward a cloud edge collaboration scheme in the field of industrial Internet, and provides a standard industrial Internet cloud edge collaboration depth integration rating standard for manufacturing enterprises, which is convenient for scientific evaluation of cloud edge synergy effect. © 2024 IEEE.

关键词： Cloud platforms

来源：评论

学校读者我要写书评

暂无评论

A reliable neural network framework for the Zika system based reservoirs and human movement

引用

Knowledge-Based Systems 2024年 292卷

作者： Sabir, Zulqurnain Khan, Sundas Naqeeb Raja, Muhammad Asif Zahoor Babatin, M.M. Hashem, Atef F. Abdelkawy, M.A. Department of Computer Science and Mathematics Lebanese American University Beirut Lebanon Department of Graphics Computer Vision and Digital Systems Silesian University of Technology Gliwice Poland Future Technology Research Center National Yunlin University of Science and Technology 123 University Road Section 3 Yunlin Douliou64002 Taiwan Riyadh11566 Saudi Arabia Department of Mathematics and Information Science Faculty of Science Beni-Suef University Egypt

The motive of current investigations provides the numerical solutions of the neuro computing solver based on the Levenberg-Marquardt backpropagation neural network approach (LMB) to solve the Zika virus system of reservoirs and human movement. The mathematical form of the human movement model is based on ten different classes, which makes the model nonlinear. The solution of this nonlinear mathematical model is presented by using the stochastic LMB neural network process. A dataset is designed based on the Adam solver in order to reduce the mean square error by dividing the data as training, endorsement, and testing with the use of 75%, 10% and 15%. Ten numbers of neurons have been taken by using the log-sigmoid transfer in the hidden layer for solving the human movement model. The precision of the proposed solver is authenticated by using the valuation of reference and proficient outcomes, while the reducible absolute error presents the correctness of the scheme. Moreover, the statistical performances based different proportional schemes have been used to authenticate the reliability of the scheme. © 2024 Elsevier B.V.

关键词： Neural networks

来源：评论

学校读者我要写书评

暂无评论

Enhancing Cyclone center Identification in Radar Images through Deep Learning and Match Recognition

Enhancing Cyclone Center Identification in Radar Images thro...

引用

International Conference on Computing and Networking Technology (ICCNT)

作者： Thirumurugan Shanmugam Chirag Chandrashekar Maheswari Subburaj Arun Kumar Sivaraman Ajmery Sultana Senthil Kumar Thangavel College of Computing and Information Sciences University of Technology and Applied Sciences Suhar Oman School of Computer Science and Engineering Vellore Institute of Technology Chennai India Digital Engineering Solution Center Photon Interactive Inc Ontario Canada School of Computer Science and Technology Algoma University Brampton Ontario Canada Department of Computer Science and Engineering Amrita Vishwa Vidyapeetham Coimbatore India

ISBN: (数字)9798350370249

ISBN: (纸本)9798350370270

This research focuses on improving the identification of cyclone centers using deep learning and match recognition applied to radar images. Accurately pinpointing the cyclone’s center is vital for predicting its intensity and trajectory. However, challenges persist in automatically locating the center due to the diverse nature of cyclone morphology and structure. To address this, the deep convolutional network’s capability is leveraged to capture various structural features in images by proposing two-step approach for cyclone center localization. Initially, a pre-trained EfficientDet model is employed using transfer learning to obtain weights. Subsequently, these weights, along with the data, are utilized in a deep learning model to provide precise coordinates of the cyclone center in the respective image. The effectiveness of existing deep learning and machine learning models show that the cyclone prediction systems have an accuracy ranging from $86 \%$ to $92 \%$ or better, and cyclone eye detection accuracy surpassing 87%. Experiment outcomes indicate that the proposed methodology outperforms conventional methods and existing works, showcasing its potential for enhancing cyclone monitoring and forecasting.

关键词： Deep learning YOLO Training Adaptation models Accuracy Machine learning algorithms Transfer learning Radar Radar imaging Cyclones

来源：评论

学校读者我要写书评

暂无评论

Visual Analysis of Point Cloud Neighborhoods via Multi-Scale Geometric Measures

Visual Analysis of Point Cloud Neighborhoods via Multi-Scale...

引用

42nd Annual Conference on European Association for computer graphics, EUROgraphics 2021

作者： Ritter, M. Schiffner, D. Harders, M. Interactive Graphics and Simulation Group The Department of Computer Science University of Innsbruck Austria DIPF Leibniz Institute for Research and Information Education Frankfurt Germany Airborne Hydromapping GmbH Innsbruck Austria

ISBN: (纸本)9783038681335

Point sets are a widely used spatial data structure in computational and observational domains, e.g. in physics particle simulations, computer graphics or remote sensing. Algorithms typically operate in local neighborhoods of point sets, for computing physical states, surface reconstructions, etc. We present a visualization technique based on multi-scale geometric features of such point clouds. We explore properties of different choices on the underlying weighted co-variance neighborhood descriptor, illustrated on different point set geometries and for varying noise levels. The impact of different weighting functions and tensor centroids, as well as point set features and noise levels becomes visible in the rotation-invariant feature images. We compare to a curvature based scale space visualization method and, finally, show how features in real-world LiDAR data can be inspected by images created with our approach in an interactive tool. In contrast to the curvature based approach, with our method line structures are highlighted over growing scales, with clear border regions to planar or spherical geometric structures. © 2021 The Author(s).

关键词： Data visualization

来源：评论

学校读者我要写书评

暂无评论

Learning Multimodal Latent Dynamics for Human-Robot Interaction

arXiv

引用

arXiv 2023年

作者： Prasad, Vignesh Heitlinger, Lea Koert, Dorothea Stock-Homburg, Ruth Peters, Jan Chalvatzaki, Georgia Department of Computer Science TU Darmstadt Germany Chair for Marketing and Human Resource Management Department of Law and Economics TU Darmstadt Germany Department of Computer Science TU Darmstadt Germany Centre for Cognitive Science TU Darmstadt Germany Germany Darmstadt Germany Interactive Robot Perception and Learning group Department of Computer Science TU Darmstadt Germany Center for Mind Brain & Behavior Uni. Marburg JLU Giessen Germany

This article presents a method for learning wellcoordinated Human-Robot Interaction (HRI) from Human- Human Interactions (HHI). We devise a hybrid approach using Hidden Markov Models (HMMs) as the latent space priors for a Variational Autoencoder to model a joint distribution over the interacting agents. We leverage the interaction dynamics learned from HHI to learn HRI and incorporate the conditional generation of robot motions from human observations into the training, thereby predicting more accurate robot trajectories. The generated robot motions are further adapted with Inverse Kinematics to ensure the desired physical proximity with a human, combining the ease of joint space learning and accurate task space reachability. For contact-rich interactions, we modulate the robot's stiffness using HMM segmentation for a compliant interaction. We verify the effectiveness of our approach deployed on a Humanoid robot via a user study. Our method generalizes well to various humans despite being trained on data from just two humans. We find that Users perceive our method as more human-like, timely, and accurate and rank our method with a higher degree of preference over other baselines. © 2023, CC BY-NC-SA.

关键词： Anthropomorphic robots

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：