检索结果-内蒙古大学图书馆

JNMR: Joint Non-Linear Motion Regression for Video Frame Interpolation

IEEE Transactions on Image Processing 2023年 32卷 5283-5295页

作者： Liu, Meiqin Xu, Chenming Yao, Chao Lin, Chunyu Zhao, Yao Beijing Jiaotong University Institute of Information Science Beijing100044 China Beijing Key Laboratory of Advanced Information Science and Network Technology Beijing100044 China University of Science and Technology Beijing School of Computer and Communication Engineering Beijing100083 China

Video frame interpolation (VFI) aims to generate predictive frames by motion-warping from bidirectional references. Most examples of VFI utilize spatiotemporal semantic information to realize motion estimation and interpolation. However, due to variable acceleration, irregular movement trajectories, and camera movement in real-world cases, they can not be sufficient to deal with non-linear middle frame estimation. In this paper, we present a reformulation of the VFI as a joint non-linear motion regression (JNMR) strategy to model the complicated inter-frame motions. Specifically, the motion trajectory between the target frame and multiple reference frames is regressed by a temporal concatenation of multi-stage quadratic models. Then, a comprehensive joint distribution is constructed to connect all temporal motions. Moreover, to reserve more contextual details for joint regression, the feature learning network is devised to explore clarified feature expressions with dense skip-connection. Later, a coarse-to-fine synthesis enhancement module is utilized to learn visual dynamics at different resolutions with multi-scale textures. The experimental VFI results show the effectiveness and significant improvement of joint motion regression over the state-of-the-art methods. The code is available at https://***/ruhig6/JNMR. © 1992-2012 IEEE.

关键词： Interpolation

来源：评论

学校读者我要写书评

暂无评论

Aperture Miniaturized Magneto-electric Dipole with Dual Linearly Polarization for 28/38 GHz Millimeter-wave Applications

Aperture Miniaturized Magneto-electric Dipole with Dual Line...

引用

2024 Photonics and Electromagnetics Research Symposium, PIERS 2024

作者： Xu, Yanhong Hu, Wanshan Wang, Zheng Li, Haoxiang Xi'an University of Science and Technology Xi'an Key Laboratory of Network Convergence Communication College of Communication and Information Engineering Xi'an710054 China

ISBN: (纸本)9798350375909

An aperture miniaturized magneto-electric (ME) dipole with dual linearly polarization (LP) is presented for 28/38 GHz millimeter-wave (mmW) on-chip applications. The aperture reduction is achieved by utilizing the self-similarity characteristic of fractal structure. In particular, four Minkowski-based structures are exploited to replace the conventional four rectangular metallic patches. An aperture of 0.4λ0 × 0.4λ0 at 33 GHz is achieved without compromising too much other excellent properties of ME dipole, such as wide impedance bandwidth, stable radiation pattern, etc. Meanwhile, in state-of-the-art antenna designs in 28/38 GHz mmW region, the majority of the studies focus on single-polarized antennas. The design of dual-polarized antenna is quite challenging. To achieve dual LP, a stripline is inserted into each of the two orthogonally placed L-shaped probes to separate the two excitation ports away from each other. The proposed antenna exhibits an overlapped 10 dB bandwidth of 45.1% ranging from 27 42 GHz and a stable gain slightly smaller than 8 dBi. A prototype based on the proposed design∼is manufactured and tested for verification. Good agreement is found between the simulated and measured results. © 2024 IEEE.

关键词： Millimeter waves

来源：评论

学校读者我要写书评

暂无评论

High Quality Audio Object Coding Framework Based on Non-Negative Matrix Factorization

引用

China communications 2017年第9期14卷 32-41页

作者： Tingzhao Wu Ruimin Hu Xiaochen Wang Shanfa Ke Jinshan Wang National Engineering Research Center for Multimedia Software School of Computer ScienceWuhan University Hubei Key Laboratory of Multimedia and Network Communication Engineering Wuhan University Collaborative Innovation Center of Geospatial Technology

Object-based audio coding is the main technique of audio scene coding. It can effectively reconstruct each object trajectory, besides provide sufficient flexibility for personalized audio scene reconstruction. So more and more attentions have been paid to the object-based audio coding. However, existing object-based techniques have poor sound quality because of low parameter frequency domain resolution. In order to achieve high quality audio object coding, we propose a new coding framework with introducing the non-negative matrix factorization(NMF) method. We extract object parameters with high resolution to improve sound quality, and apply NMF method to parameter coding to reduce the high bitrate caused by high resolution. And the experimental results have shown that the proposed framework can improve the coding quality by 25%, so it can provide a better solution to encode audio scene in a more flexible and higher quality way.

关键词： object-based audio coding non-negative matrix factorization audio scenecoding

来源：评论

学校读者我要写书评

暂无评论

Perovskite light-emitting/detecting bifunctional fibres for wearable LiFi communication

引用

Light(science & Applications) 2020年第1期9卷 441-449页

作者： Qingsong Shan Changting Wei Yan Jiang Jizhong Song Yousheng Zou Leimeng Xu Tao Fang Tiantian Wang Yuhui Dong Jiaxin Liu Boning Han Fengjuan Zhang Jiawei Chen Yongjin Wang Haibo Zeng MIIT Key Laboratory of Advanced Display Materials and Devices Institute of Optoelectronics&NanomaterialsSchool of Materials Science and EngineeringNanjing University of Science and TechnologyNanjing 210094China Peter Grünberg Research Center Laboratory of Broadband Wireless Communication and Sensor Network TechnologyMinistry of EducationNanjing University of Posts and TelecommunicationsNanjing 210003China

Light fidelity(LiFi),which is emerging as a compelling technology paradigm shifting the common means of highcapacity wireless communication technologies,requires wearable and full-duplex compact design because of its great significance in smart wearables as well as the‘Internet of Things’.However,the construction of the key component of wearable full-duplex LiFi,light-emitting/detecting bifunctional fibres,is still challenging because of the conflicting process between carrier separation and recombination,as well as the highly dynamic film-forming ***,we demonstrate light-emitting/detecting bifunctional fibres enabled by perovskite QDs with hybrid *** hybrid perovskite inks endow fibres with super-smooth QD ***,combined with the small exciton binding energy and high carrier mobility of perovskite QDs,enables successful integration of electroluminescence and photodetection into *** bifunctional fibres possess the narrowest electroluminescence full width at half maximum of ~19 nm and,more importantly,the capability for simultaneously transmitting and receiving *** successful fabrication of narrow emission full-duplex LiFi fibres paves the way for the fabrication and integration of low crosstalk interoperable smart wearables.

关键词： process perovskite functional

来源：评论

学校读者我要写书评

暂无评论

Learning Modality-Invariant Feature for Multimodal Image Matching via Knowledge Distillation

引用

IEEE Transactions on Geoscience and Remote Sensing 2025年 63卷

作者： Yepeng Liu Wenpeng Lai Yuliang Gu Gui-Song Xia Bo Du Yongchao Xu National Engineering Research Center for Multimedia Software Institute of Artificial Intelligence School of Computer Science and Hubei Key Laboratory of Multimedia and Network Communication Engineering Wuhan University Wuhan China SF Technology Company Ltd. Shenzhen China

Multimodal remote sensing image matching is essential for multisource information fusion. Recently, learning-based feature matching networks have significantly enhanced the performance of unimodal image matching tasks through data-driven approaches. However, progress in applying these learning-based methods to multimodal image matching has been slower. A major obstacle is the substantial nonlinear radiometric differences between modalities, which require networks to learn modality-invariant features from large amounts of paired data. To address this, we propose EMINet, an efficient method for learning modality-invariant features from limited data to improve matching performance. Our approach constructs a high-performance teacher network by combining the DINOv2 foundational model, the keypoint and descriptor extraction network SuperPoint, and the feature matching network SuperGlue. Leveraging the strong semantic representation capability of DINOv2, the teacher network achieves excellent cross-modality matching ability. To meet low-latency requirements in practical applications, we introduce two novel knowledge distillation strategies: semantic window relation distillation (SWRD) and cross-triplet descriptor distillation (CTDD). SWRD improves the discriminative power of the student network’s descriptors by learning patch-level distributions from DINOv2, while CTDD enforces cross-modality triplet constraints to enhance modality invariance of the student network. Experimental results demonstrate that EMINet outperforms several state-of-the-art methods on various datasets, including optical-SAR, optical-NIR, and optical-IR datasets.

关键词： Image matching Feature extraction Remote sensing Optical sensors Optical imaging Semantics Radiometry Nonlinear optics Adaptive optics Training

来源：评论

学校读者我要写书评

暂无评论

Scalable load balancing for large-scale web server clusters

引用

Journal of Computational Information Systems 2014年第13期10卷 5763-5771页

作者： Huang, Chuibi Wang, Jinlin Wu, Gang Chen, Jun Department of Automation University of Science and Technology of China Key Laboratory of Network Communication System and Control Hefei China Institute of Acoustics Chinese Academy of Sciences National Network New Media Engineering Research Center Beijing China

Load balancing is a fundamental technology to provide scalability for web server clusters. But in the scenarios where the web server clusters have massive amount of user accesses, a single load balancing server is unable to carry the user visits. Therefore load balancing tasks should be undertaken among a set of load balancing servers. Correspondingly the loads among these servers also need to be balanced. In this paper, we proposed a load balancing scheme that could provide scalability, availability and reliability for large-scale web server clusters. The scheme combines existing routing techniques and splits the function of load balancing into three levels that could be implemented by different devices, thereby enabling the web server cluster scales with user accesses. The experiment results have demonstrated the effectiveness of the program. © 2014 by Binary Information Press

关键词： Scalability

来源：评论

学校读者我要写书评

暂无评论

Joint Optimization of Latency and Energy Consumption for Mobile Edge Computing Based Proximity Detection in Road networks

引用

China communications 2022年第4期19卷 274-290页

作者： Tongyu Zhao Yaqiong Liu Guochu Shou Xinwei Yao School of Information and Communication Engineering Beijing University of Posts and TelecommunicationsBeijing 100876China Beijing Laboratory of Advanced Information Networks Beijing Key Laboratory of Network System Architecture and ConvergenceBeijing 100876China School of Computer Science and Technology Zhejiang University of TechnologyChina

In recent years, artificial intelligence and automotive industry have developed rapidly, and autonomous driving has gradually become the focus of the industry. In road networks, the problem of proximity detection refers to detecting whether two moving objects are close to each other or not in real time. However, the battery life and computing capability of mobile devices are limited in the actual scene,which results in high latency and energy consumption. Therefore, it is a tough problem to determine the proximity relationship between mobile users with low latency and energy consumption. In this article, we aim at finding a tradeoff between latency and energy consumption. We formalize the computation offloading problem base on mobile edge computing(MEC)into a constrained multiobjective optimization problem(CMOP) and utilize NSGA-II to solve it. The simulation results demonstrate that NSGA-II can find the Pareto set, which reduces the latency and energy consumption effectively. In addition, a large number of solutions provided by the Pareto set give us more choices of the offloading decision according to the actual situation.

关键词： proximity detection mobile edge computing road networks constrained multiobjective optimization

来源：评论

学校读者我要写书评

暂无评论

A Novel 3D Video Fingerprinting Algorithm Based on Local Feature Points

引用

Chinese Journal of Electronics 2018年第6期27卷 1192-1199页

作者： SUN Ziqiang ZHU Yuesheng LIU Xiyao Shenzhen Key Lab of Information Theory & Future Network Arch Communication and Information Security LaboratoryShenzhen Graduate School Peking University Nanjing Research Institute of Electronics Technology School of Information Science and Engineering Central South University

Digital rights management of the 3 D contents is a crucial open issue in the 3 D video industry. A novel robust fingerprinting algorithm is proposed for protecting the copyright of the 3 D video. Unlike the existing algorithms extracting visual features separately from the 2 D videos and the depth maps, in our algorithm a novel local stereo space is constructed according to the depth information of the pixels around the extracted local feature points in the 2 D videos, and the 3 D videos are processed in a holistic manner. In the proposed space,the 3 D-transform-feature is extracted and aggregated into a feature matrix, and then the compact 3 D video fingerprints are obtained from the eigenspace of the matrix. Our comprehensive experiments are conducted on a 3 D video database, and the results have demonstrated the robustness and discrimination of the proposed algorithm. Moreover, our fingerprints cost less storage spaces than the existing approaches.

关键词： Digital rights management Video fingerprinting 3D video Local feature points

来源：评论

学校读者我要写书评

暂无评论

Hybrid entanglement concentration assisted with single coherent state

引用

Chinese Physics B 2016年第3期25卷 65-72页

作者：郭锐周澜顾世浦王兴福盛宇波 Key Laboratory of Broadband Wireless Communication and Sensor Network Technology (Ministry of Education) Nanjing University of Posts and Telecommunications College of Mathematics & Physics Nanjing University of Posts and Telecommunications College of Electronic Science and Engineering Nanjing University of Posts and Telecommunications

Hybrid entangled state （HES） is a new type of entanglement, which combines the advantages of an entangled po- larization state and an entangled coherent state. HES is widely discussed in the applications of quantum communication and computation. In this paper, we propose three entanglement concentration protocols （ECPs） for Bell-type HES, W-type HES, and cluster-type HES, respectively. After performing these ECPs, we can obtain the maximally entangled HES with some success probability. All the ECPs exploit the single coherent state to complete the concentration. These protocols are based on the linear optics, which are feasible in future experiments.

关键词： hybrid entangled state quantum communication and computation entanglement concentration

来源：评论

学校读者我要写书评

暂无评论

Stochastic prefix-based fault tolerant source routing algorithm for network on chip

Stochastic prefix-based fault tolerant source routing algori...

引用

2012 International Conference on Computer and Information science, ICCIS 2012 - A Conference of World Engineering, science and technology Congress, ESTCON 2012

作者： Xu, Gengchun Ren, Kaixin Gu, Naijie Anhui Province Key Laboratory of Computing and Communication Software Hefei China USTC-SICT Joint Laboratory on Network and Communication Hefei China School of Computer Science and Technology University of Science and Technology of China Hefei China

ISBN: (纸本)9781467319386

In this paper, a novel fault-tolerant source routing algorithm for network on Chip (NOC) called Stochastic Prefix-based Source Routing (SPSR) is proposed. Compared to non-source routing algorithms, it releases the intermediate node from complicated decision-making or routing table maintenance. Unlike the traditional source routing algorithms which use broadcast packets to record the YD of every intermediate node in Route Discovery, it predetermine prospective routes, which can greatly reduces the packets generated and forwarded in the Route Discovery phase, and therefore reduces the power consumption. The proposed routing algorithm is implemented on the open source simulator Nirgam [14], incorporating Orion [15] to estimate the power consumption. Experiment results show that the proposed algorithm can greatly reduce the power consumption in the route discovery and the latency of SPSR is comparable with that of the broadcasting mechanism. © 2012 IEEE.

关键词： network-on-chip

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：