检索结果-内蒙古大学图书馆

Conference on computer Vision and Pattern Recognition (CVPR)

作者： Jin Lin Xiaotong Luo Ming Hong Yanyun Qu Yuan Xie Zongze Wu School of Informatics Xiamen University Fujian China School of Computer Science and Technology East China Normal University Shanghai China School of Mechatronics and Control Engineering Shenzhen University Shenzhen China

Scalable deep Super-Resolution (SR) models are increasingly in demand, whose memory can be customized and tuned to the computational recourse of the platform. The existing dynamic scalable SR methods are not memory-friendly enough because multi-scale models have to be saved with a fixed size for each model. Inspired by the success of Lottery Tickets Hypothesis (LTH) on image classification, we explore the existence of unstructured scalable SR deep models, that is, we find gradual shrinkage subnetworks of extreme sparsity named winning tickets. In this paper, we propose a Memory-friendly Scalable SR framework (MSSR). The advantage is that only a single scalable model covers multiple SR models with different sizes, instead of reloading SR models of different sizes. Concretely, MSSR consists of the forward and backward stages, the former for model compression and the latter for model expansion. In the forward stage, we take advantage of LTH with rewinding weights to progressively shrink the SR model and the pruning-out masks that form nested sets. Moreover, stochastic self-distillation (SSD) is conducted to boost the performance of sub-networks. By stochastically selecting multiple depths, the current model inputs the selected features into the corresponding parts in the larger model and improves the performance of the current model based on the feedback results of the larger model. In the backward stage, the smaller SR model could be expanded by recovering and fine-tuning the pruned parameters according to the pruning-out masks obtained in the forward. Extensive experiments show the effectiveness of MMSR. The smallest-scale sub-network could achieve the sparsity of 94% and outperforms the compared lightweight SR methods.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Application of Structured-light 3D Reconstruction in Virtual Assembly 3

Application of Structured-light 3D Reconstruction in Virtual...

引用

2021年第三届先进计算机技术、信息科学与通信国际会议(CTISC2021)

作者： Youteng Wan Ang Cai Qi Wei Jing Liu Xiangjia Chen Jiaming Feng Zhan Song Juan Zhao Shenzhen Institutes of Advanced Technology Chinese Academy of Sciences Institute of Laser Engineering Faculty of Materials and Manufacturing Beijing University of Technology School of Physics and Optoelectronics Faculty of Science Beijing University of Technology

ISBN: (纸本)9781665418683

3D modeling is a time-consuming and laborious work in virtual assembly. How to reconstruct parts with various shapes quickly and accurately is an important topic in the field of reverse engineering. In this paper, a popgun and its special-shaped parts are reconstructed by using a selfdeveloped surface structured light scanning system and a commercial line structured light scanning system, with an accuracy less than 0.1 mm. And the virtual assembly of the popgun is completed according to the high-precision 3D solid models.

关键词： Solid modeling Analytical models Surface reconstruction Three-dimensional displays Shape Reverse engineering Solids

来源：评论

学校读者我要写书评

暂无评论

A Numerical solution of the linear elasticity problem using the FEM 2

A Numerical solution of the linear elasticity problem using ...

引用

2nd IEEE International Conference on Electronic engineering, ICEEM 2021

作者： Abdelhamid, Talaat Chen, Rongliang Alam, Md. Mahbub Chinese Academy of Sciences Shenzhen Institutes of Advanced Technology Shenzhen518055 China Physics and Mathematical Engineering Department Faculty of Electronic Engineering Menoufiya University Menouf Egypt Harbin Institute of Technology Center for Turbulence and Control Shenzhen518055 China

ISBN: (纸本)9781665418423

The numerical solution of the 2D linear elasticity problem numerically from the given body and traction forces, Poisson ratio, and Young's modulus is investigated. A continuous linear finite element method is adapted to solve the forward elasticity problem when Young's modulus is defined as variable valued function. The body's deformation is approximated by applying an approximation of the plane stress condition. We have studied the convergence analysis of the proposed linear elasticity problem. A numerical example presents the worthiness of the proposed technique. © 2021 IEEE.

关键词： Finite element method

来源：评论

学校读者我要写书评

暂无评论

Vertical Take-Off and Landing fixed-wing designed for autonomous missions

Vertical Take-Off and Landing fixed-wing designed for autono...

引用

IEEE/ASME (AIM) International Conference on advanced Intelligent Mechatronics

作者： Krzysztof Lewandowski Jakub Tomczak Jakub Zeifert Szymon Nowacki Marcel Król Dawid Rudy Jacek Grzybowski Paweł Piórkowski Roman Czyba Marcin Lemanowicz Piotr Czekalski Department of Automatic Control Faculty of Automatic Control Electronics and Computer Science Silesian University of Technology Gliwice Poland Department of Chemical Engineering and Process Design Faculty of Chemistry Silesian University of Technology Gliwice Poland

This paper presents the design and development of a Vertical Take-off and Landing (VTOL) fixed-wing aircraft intended for autonomous missions. It provides an overview of the current state of VTOL technology and its applications. The paper focuses on fixed-wing VTOL aircraft created by Academic Scientific Association High Flyers from the Silesian university of technology in Poland. The design process and considerations are discussed in detail, including aerodynamics, selection of materials, hardware, control systems and software. Finally, the paper discusses real-world scenarios that the designed UAV could be used to solve real-world problems, such as targeted plant protection or the deployment of oral vaccines for wildlife. The authors successfully tested solutions presented in the paper during competitions and real practical applications. Overall, this paper provides a comprehensive look into the design and development of a VTOL aircraft for autonomous missions and presents its effectiveness and capabilities in solving real-life problems.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Multimodal Large Models Are Effective Action Anticipators

arXiv

引用

arXiv 2025年

作者： Wang, Binglu Tian, Yao Wang, Shunzhou Yang, Le College of Information and Control Engineering Xi'an University of Architecture and Technology Xi'An710055 China The School of Astronautics Northwestern Polytechnical University Xi'An710072 China School of Electronic and Computer Engineering Shenzhen Graduate School Peking University Shenzhen518055 China Institute of Artificial Intelligence Hefei Comprehensive National Science Center China The School of Information Science and Technology University of Science and Technology of China Hefei230022 China

The task of long-term action anticipation demands solutions that can effectively model temporal dynamics over extended periods while deeply understanding the inherent semantics of actions. Traditional approaches, which primarily rely on recurrent units or Transformer layers to capture long-term dependencies, often fall short in addressing these challenges. Large Language Models (LLMs), with their robust sequential modeling capabilities and extensive commonsense knowledge, present new opportunities for long-term action anticipation. In this work, we introduce the ActionLLM framework, a novel approach that treats video sequences as successive tokens, leveraging LLMs to anticipate future actions. Our baseline model simplifies the LLM architecture by setting future tokens, incorporating an action tuning module, and reducing the textual decoder layer to a linear layer, enabling straightforward action prediction without the need for complex instructions or redundant descriptions. To further harness the commonsense reasoning of LLMs, we predict action categories for observed frames and use sequential textual clues to guide semantic understanding. In addition, we introduce a Cross-Modality Interaction Block, designed to explore the specificity within each modality and capture interactions between vision and textual modalities, thereby enhancing multimodal tuning. Extensive experiments on benchmark datasets demonstrate the superiority of the proposed ActionLLM framework, encouraging a promising direction to explore LLMs in the context of action anticipation. Code is available at https://***/2tianyao1/***. Copyright © 2025, The Authors. All rights reserved.

关键词： Semantics

来源：评论

学校读者我要写书评

暂无评论

MULSAM: Multidimensional Attention with Hardware Acceleration for Efficient Intrusion Detection on Vehicular CAN Bus

引用

IEEE Transactions on computer-Aided Design of Integrated Circuits and Systems 2025年

作者： Xu, He Shi, Xiaokang Liu, Hansheng Wang, Yanwen Lu, Jiwu Zeng, Haibo Li, Renfa Wu, Di Hunan University National Engineering Research Center for Robot Visual Perception and Control Technology Hunan Changsha410082 China Hunan University College of Electrical and Information Engineering Hunan Changsha410082 China Hunan University Shenzhen Research Institute Shenzhen518000 China Virginia Tech Department of Electrical and Computer Engineering BlacksburgVA24061 United States Hunan University College of Computer Science and Electronic Engineering Hunan Changsha410082 China

controller Area Network (CAN) protocol is an efficient standard enabling communication among Electronic control Units (ECUs). However, the CAN bus is vulnerable to malicious attacks because of a lack of defense features. In this article, a novel vehicle Intrusion Detection System (IDS) is developed. The challenge is that existing techniques of Intrusion Detection Systems (IDS) rarely consider attacks with small-batch, which are characterized by their small attack scale and concealed attack patterns, posing a significant threat to driving safety. To solve this problem, we developed an algorithm model that merges Multi-dimensional Long Short-Term Memory (MD-LSTM) and Self-Attention Mechanism (SAM), shortly named MULSAM. The MULSAM model was compared with other baseline models, including Stacked LSTM, MD-LSTM, etc. Experiments show that our approach has the best detection accuracy (98.98%) and training stability. Further, to speed up the inference of MULSAM on edge, the hardware accelerator is implemented on FPGA devices using technologies such as parallelization, modular, pipeline, and fixed-point quantization. Experiments show that our FPGA-based acceleration scheme has a better energy efficiency than the CPU platform. Even with a certain degree of quantification, the acceleration model for MULSAM still displays a high detection accuracy of 98.81% and a low latency of 1.88 ms. © 1982-2012 IEEE.

关键词： Intrusion detection

来源：评论

学校读者我要写书评

暂无评论

RISAC: Rate-splitting Multiple Access Enabled Integrated Sensing and Communication Systems

RISAC: Rate-splitting Multiple Access Enabled Integrated Sen...

引用

IEEE International Conference on Communications (ICC)

作者： Zheng Liu Yuxuan Jint Bin Cao Rongxing Lu School of Electronics and Information Engineering Harbin Institute of Technology Shenzhen China Faculty of Computer Science University of New Brunswick Fredericton Canada

Integrated sensing and communication (ISAC) is considered to be a promising paradigm for future wireless evolution and emerging services. In such a framework, the spectrum efficiency and hardware utilization are significantly enhanced, since the radar detection and data transmission share the same spectrum and hardware. This paper studies a rate-splitting multiple access enabled ISAC (RISAC), wherein the multi-antenna RISAC transmitter simultaneously senses a single target with the help of a multi-antenna radar receiver, and communicates with a group of users with single-antenna for data transmission based on RSMA. We focus on optimizing the sensing performance in terms of Cramér-Rao lower bound (CRLB), while achieving acceptable data transmission performance in terms of data rate. To this end, we aim to minimize the CRLB of the sensing response matrix at the radar receiver, by adequately designing the RSMA structure and the associated parameters, with constraints of data rate requirement and transmit power budget. To tackle the non-convex CRLB minimization problem, we split the original problem into outer and inner subproblems, which can be efficiently solved by particle swarm optimization algorithm and semidefinite relaxation method, respectively. Numerical results confirm that RISAC outperforms the space division multiple access based strategy in terms of CRLB.

关键词：

来源：评论

学校读者我要写书评

暂无评论

AI in drug development: advances in response, combination therapy, repositioning, and molecular design

引用

science CHINA Information sciences 2025年第7期68卷 1-17页

作者： Ren QI Shujia LIU Xingqi HUI Alexey K. SHAYTAN Bin LIU School of Computer Science and Technology Beijing Institute of Technology Beijing 100081 China Department of Biology Lomonosov Moscow State University Moscow 119234 Russia International Laboratory of Bioinformatics AI and Digital Sciences Institute Faculty of Computer Science HSE University Moscow 109028 Russia SMBU-MSU-BIT Joint Laboratory on Bioinformatics and Engineering Biology Shenzhen MSU-BIT University Shenzhen 518172 China Zhongguancun Academy Beijing 100094 China

Artificial intelligence (AI) is revolutionizing the field of drug development, particularly in addressing key challenges such as drug response prediction, drug combination design, drug repositioning, and drug molecule generation. Traditional drug discovery is hindered by long timelines, high costs, and low success rates, necessitating innovative technologies to accelerate the process. AI technologies, such as deep learning, graph neural networks, and generative models, have demonstrated significant potential in enhancing the accuracy of drug response predictions, optimizing drug combination strategies, identifying opportunities for drug repositioning, and generating drug molecules with specific biological activities. These advancements not only accelerate the drug development process but also open up new possibilities for precision medicine. This review discusses the latest applications and developments of AI in drug discovery, highlighting the breakthroughs and challenges AI addresses in drug development. By summarizing the current research progress, this study provides theoretical support and practical guidance for further applications of AI in drug development.

关键词： artificial intelligence deep learning drug research drug repositioning genomic data

来源：评论

学校读者我要写书评

暂无评论

Multi-modal Multi-kernel Graph Learning for Autism Prediction and Biomarker Discovery

arXiv

引用

arXiv 2023年

作者： Liu, Jin Mao, Junbin Lin, Hanhe Kuang, Hulin Pan, Shirui Wu, Xusheng Xie, Shan Liu, Fei Pan, Yi Hunan Provincial Key Lab on Bioinformatics School of Computer Science and Engineering Central South University Changsha410083 China Xinjiang Engineering Research Center of Big Data and Intelligent Software School of software Xinjiang University Wulumuqi830000 China Hunan Province Key Lab on Bioinformatics School of Computer Science and Engineering Central South University Changsha410083 China Faculty of Computer Science and Control Engineering Shenzhen University of Advanced Technology Shenzhen Institute of Advanced Technology Chinese Academy of Sciences Shenzhen518055 China School of Science and Engineering University of Dundee DundeeDD1 4HN United Kingdom School of Information and Communication Technology Griffith University Gold CoastQLD4215 Australia Shenzhen Health Development Research and Data Management Center Shenzhen518109 China

Graph learning-based multi-modal integration and classification is one of the most challenging tasks for disease prediction. To effectively offset the negative impact among modalities in the process of multi-modal integration and heterogeneous information extractions from graphs, we propose a novel method called Multi-modal Multi-Kernel Graph Learning (MMKGL). To solve the problem of negative impact among modalities, we propose a multi-modal graph embedding module to construct a multi-modal graph. Different from conventional methods that manually construct static graphs for all modalities, each modality generates a separate graph by adaptive learning, where a function graph and a supervision graph are introduced for optimization during the multi-graph fusion embedding process. We then propose a multi-kernel graph learning module to extract heterogeneous information from the multi-modal graph. The information in the multi-modal graph at different levels is aggregated by convolutional kernels with different receptive field sizes, followed by generating a cross-kernel discovery tensor for disease prediction. Our method is evaluated on the benchmark Autism Brain Imaging Data Exchange (ABIDE) dataset and outperforms the state-of-the-art methods. In addition, discriminative brain regions associated with autism are identified by our model, providing guidance for the study of autism pathology. The source code will be available at https://***/yutian0315/MMKGL. © 2023, CC BY-NC-ND.

关键词： Graph embeddings

来源：评论

学校读者我要写书评

暂无评论

Size-Invariance Matters: Rethinking Metrics and Losses for Imbalanced Multi-object Salient Object Detection 41

Size-Invariance Matters: Rethinking Metrics and Losses for I...

引用

41st International Conference on Machine Learning, ICML 2024

作者： Li, Feiran Xu, Qianqian Bao, Shilong Yang, Zhiyong Cong, Runmin Cao, Xiaochun Huang, Qingming Institute of Information Engineering Chinese Academy of Sciences Beijing China School of Cyber Security University of Chinese Academy of Sciences Beijing China Key Laboratory of Intelligent Information Processing Institute of Computing Technology Chinese Academy of Sciences Beijing China School of Computer Science and Technology University of Chinese Academy of Sciences Beijing China Institute of Information Science Beijing Jiaotong University Beijing China School of Control Science and Engineering Shandong University Jinan China Key Laboratory of Machine Intelligence and System Control Ministry of Education Jinan China School of Cyber Science and Tech. Sun Yat-sen University Shenzhen Campus China Key Laboratory of Big Data Mining and Knowledge Management Chinese Academy of Sciences Beijing China

This paper explores the size-invariance of evaluation metrics in Salient Object Detection (SOD), especially when multiple targets of diverse sizes co-exist in the same image. We observe that current metrics are size-sensitive, where larger objects are focused, and smaller ones tend to be ignored. We argue that the evaluation should be size-invariant because bias based on size is unjustified without additional semantic information. In pursuit of this, we propose a generic approach that evaluates each salient object separately and then combines the results, effectively alleviating the imbalance. We further develop an optimization framework tailored to this goal, achieving considerable improvements in detecting objects of different sizes. Theoretically, we provide evidence supporting the validity of our new metrics and present the generalization analysis of SOD. Extensive experiments demonstrate the effectiveness of our method. The code is available at https://***/Ferry-Li/SI-SOD. Copyright 2024 by the author(s)

关键词： Object detection

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：