检索结果-内蒙古大学图书馆

8th IEEE International Conference on Smart Cloud, SmartCloud 2023

作者： Yuan, Gaojing Huang, Weihong Cai, Jiahong School of Computer Science and Engineering Hunan University of Science and Technology Xiangtan China

ISBN: (纸本)9798350313505

Target tracking has also received a great deal of attention from society. The basic problem of target tracking algorithm select target of interest video or image sequence, and in the next consecutive frames, find the exact position of that target and form its motion trajectory. In the early days of dealing with target tracking the optical flow method (Lucas-Kanade), Kalman filtering, kernel methods, etc. were usually used to achieve target tracking. But since 2016, deep learning has also made breakthroughs target tracking as well as target detection. In this paper, we introduce the traditional target tracking algorithms and then describe and discuss in detail the target tracking algorithms around deep learning, to explore the excellence of deep learning in target tracking compared with the earlier tracking algorithms as well as the broad prospects for development in the future. ©2023 IEEE.

关键词： Target tracking

来源：评论

学校读者我要写书评

暂无评论

Advancing Matrix Decomposition Efficiency: A Study on FT-Matrix DSP Based SVD Optimization 10

Advancing Matrix Decomposition Efficiency: A Study on FT-Mat...

引用

10th IEEE International Conference on Cyber Security and Cloud Computing and 9th IEEE International Conference on Edge Computing and Scalable Cloud, CSCloud-EdgeCom 2023

作者： Xie, Anxing Hu, Yonghua Cheng, Aobo Tang, Zhuoyou Liu, Peng Zhang, Xin Hunan University of Science and Technology School of Computer Science and Engineering Xiangtan China

ISBN: (纸本)9798350312461

Matrix decomposition is a fundamental operation in linear algebra, and it has various applications in machine learning, signal processing, edge computing, and many other fields. Singular Value Decomposition (SVD) is a matrix decomposition method that can break down a matrix into three matrices: two orthogonal matrices and a diagonal matrix. With the development of domestic high-performance Digital Signal Value Processors (DSP), the demand for matrix computation based on DSP platforms is increasing. The research of SVD implemented based on DSP is important and meaningful. However, accessing the high-performance algorithm requires developers who are familiar with the hardware characteristics, in order to combine the unique features of the algorithm with the limited hardware resources. To reduce the cost of computing the SVD in matrix, we implement a vectorization mapping method for the SVD algorithm on the FT-M7002. The single instruction multiple data (SIMD) instructions embedded in the FT-M7002 processor were utilized to exploit the data-level parallelism in the SVD algorithm. Instead of using data movement and a scalar processing unit (SPU), we compute with a single vector processing element (VPE). Additionally, DMA transfer algorithm is designed to implement matrix transposition and resolve the issue of discontinuous data access. Experimental results show that the optimized SVD algorithm improves execution performance relative to the original SVD algorithm on FT by up to 5.0 ×. Furthermore, we demonstrate that the optimized SVD algorithm on the FT-M7002 performs 1.0-2.0× faster than the optimized SVD algorithm on TMS320C6678 processor. © 2023 IEEE.

关键词： Singular value decomposition

来源：评论

学校读者我要写书评

暂无评论

Toward actionable testing of deep learning models

引用

science China(Information sciences) 2023年第7期66卷 300-302页

作者： Yingfei XIONG Yongqiang TIAN Yepang LIU Shing-Chi CHEUNG Key Laboratory of High Confidence Software Technologies(Peking University) Ministry of Education School of Computer Science Peking University Department of Computer Science and Engineering The Hong Kong University of Science and Technology School of Computer Science University of Waterloo Research Institute of Trustworthy Autonomous Systems Southern University of Science and Technology Department of Computer Science and Engineering Southern University of Science and Technology

Deep learning has become an important computational paradigm in our daily lives with a wide range of applications,from authentication using facial recognition to autonomous driving in smart vehicles. The quality of the deep learning models, i.e., neural architectures with parameters trained over a dataset, is crucial to our daily living and economy.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Auto-3D-house Design from Structured User Requirements

引用

Machine Intelligence Research 2025年第2期22卷 368-385页

作者： Minkui Tan Qi Chen Zixiong Huang Qi Wu Yuanqing Li Jiaqiu Zhou School of Software Engineering South China University of TechnologyGuangzhou 510006China Faculty of Engineering Computer and Mathematical ScienceUniversity of AdelaideAdelaide SA 5005Australia School of Automation Science and Engineering South China University of TechnologyGuangzhou 510006China

We study the task of automated house design,which aims to automatically generate 3D houses from user ***,in the automatic system,it is non-trivial due to the intrinsic complexity of house designing:1)the understanding of user requirements,where the users can hardly provide high-quality requirements without any professional knowledge;2)the design of house plan,which mainly focuses on how to capture the effective information from user *** address the above issues,we propose an automatic house design framework,called auto-3D-house design(A3HD).Unlike the previous works that consider the user requirements in an unstructured way(e.g.,natural language),we carefully design a structured list that divides the requirements into three parts(i.e.,layout,outline,and style),which focus on the attributes of rooms,the outline of the building,and the style of decoration,*** the processing of architects,we construct a bubble diagram(i.e.,graph)that covers the rooms′attributes and relations under the constraint of *** addition,we take each outline as a combination of points and orders,ensuring that it can represent the outlines with arbitrary ***,we propose a graph feature generation module(GFGM)to capture layout features from the bubble diagrams and an outline feature generation module(OFGM)for outline ***,we render 3D houses according to the given style requirements in a rule-based *** on two benchmark datasets(i.e.,RPLAN and T3HM)demonstrate the effectiveness of our A3HD in terms of both quantitative and qualitative evaluation metrics.

关键词： Automated house design user requirements understanding outline processing layout generation graph feature generation.

来源：评论

学校读者我要写书评

暂无评论

An Iterative Neighborhood Blending Method For Multimodal Product Retrieval 2

An Iterative Neighborhood Blending Method For Multimodal Pro...

引用

2nd Asia Conference on Advanced Robotics, Automation, and Control engineering, ARACE 2023

作者： Liu, Qixin He, Tingqin Zhou, Zisheng Hunan University of Science and Technology School of Computer Science and Engineering Xiangtan China

ISBN: (纸本)9798350328363

Recently, the product matching and retrieval methods based on users' real needs are becoming a hot spot in academic research, especially in the era of big data, more efficient and accurate matching methods become more necessary. Therefore, this paper aims to design an iterative neighborhood blending method for multimodal product retrieval, this method fuses image and text features of products that can efficiently retrieve items of interest to users from massive product data. First, the feature extraction model is trained by the classification head in the training stage. secondly, post-processing methods such as pooling, batch normalization and feature normalization method are designed to obtain feature vectors of different dimensions. Thirdly, product matching and retrieval are realized by an improved iterative neighborhood mixing method. Finally, experiments on the Shopee-Price dataset demonstrate that the iterative neighborhood blending algorithm brings significant improvements to the multimodal product retrieval task compared to the state-of-art. © 2023 IEEE.

关键词： Iterative methods

来源：评论

学校读者我要写书评

暂无评论

Revolutionizing Image Captioning: Integrating Attention Mechanisms with Adaptive Fusion Gates

IAENG International Journal of Computer Science

引用

IAENG International Journal of computer science 2024年第3期51卷 212-221页

作者： Sheng, Shou-Jun Zhou, Zi-Wei School of Computer and Software Engineering University of Science and Technology Liaoning AnShan114051 China

In order to dynamically create a sequence of textual descriptions for images, image description models often make use of the attention mechanism, which involves an automatic focus on different regions within an image. However, a prevalent issue with current attention mechanisms is their tendency to overlook essential elements within the image, prioritizing contextual aspects of the object when generating descriptive text. This constraint results in a decrease in the precision of the textual descriptions produced. To address this issue and improve the accuracy of image interpretation, a proposed model for image description utilizes an attention-based approach and includes a multi-layer decoder and a fusion gate. This model is based on an encoder-decoder architecture and utilizes Residual Network (ResNet) framework for feature extraction during the encoding phase, thereby extending the encoder-decoder structure into the decoding phase. Within this framework, an adaptive fusion gate mechanism is introduced and combined with multi-layer cascade decoders to facilitate the generation of utterances. This allows decoders from lower layers to actively contribute to the final text prediction phase, thereby incrementally improving the accuracy of the predicted text and generating more precise descriptions. MS COCO 2014 dataset has been utilized for the purpose of training and validating the effectiveness of this model in understanding images. The results clearly and unequivocally establish the model’s capacity to generate exceptional predictions. When compared with the top-performing models, it has demonstrated significant enhancements, as indicated by a 0.096 rise in BLEU_1 metric, a 0.153 improvement in ROUGE_L metric, and a remarkable 0.32 increase in CIDEr metric on MS COCO dataset. The overall improvement in performance across all evaluation criteria highlights the model’s alignment with the requirements of image understanding applications. © (2024), (International Associ

关键词： computer vision

来源：评论

学校读者我要写书评

暂无评论

Performances,mechanisms,and strategies for methane removal via gas biofiltration

引用

Frontiers of Environmental science & engineering 2025年第5期19卷 141-162页

作者： Xin Wu Chunping Yang Yan Lin Xiang Li Shaohua Wu Piotr Rybarczyk Jacek Gebicki Academy of Environmental and Resource Sciences School of Environmental Science and EngineeringGuangdong University of Petrochemical TechnologyMaoming 525000China College of Urban and Environment Sciences Hunan Provincial Key Laboratory of Comprehensive Utilization of Agricultural and Animal Husbandry Waste ResourcesHunan University of TechnologyZhuzhou 412007China College of Environmental Science and Engineering Hunan University and Key Laboratory of Environmental Biology and Pollution Control(Hunan University)of Ministry of EducationChangsha 410082China School of Environmental Science and Engineering Hainan UniversityHaikou 570228China College of Chemistry and Chemical Engineering Hunan Normal UniversityChangsha 410081China Department of Process Engineering and Chemical Technology Faculty of ChemistryGdansk University of TechnologyGdansk 80-233Poland

Control of methane emissions has attracted widespread attention in the context of worldwide efforts to alleviate the greenhouse ***,a green and cost-effective treatment process,has been widely used for the purification of ***,there is still a lack of comprehensive information on methane ***,the state-of-the-art development trends of methane removal via biofiltration were comprehensively reviewed and commented *** of all,this article reviewed the mechanisms of methane removal and the possible degradation pathways in ***,the removal characteristics of single methane,methane mixed with volatile organic compounds or gaseous inorganic compounds by biofiltration,were summarized and ***,the practical applications of methane biofiltration were summed up and some inspirations were ***,the effective measures to improve the performance of methane biofiltration were put forward,and corresponding insights and future research directions for methane biofiltration were *** article can provide references for the practical applications of methane biofiltration and the mitigation of greenhouse effects.

关键词： Biofilter Methane VOC Greenhouse effect Carbon reduction

来源：评论

学校读者我要写书评

暂无评论

A New Multi-Feature Recommendation Model Based on Recurrent Neural Network 10

A New Multi-Feature Recommendation Model Based on Recurrent ...

引用

10th IEEE International Conference on Cyber Security and Cloud Computing and 9th IEEE International Conference on Edge Computing and Scalable Cloud, CSCloud-EdgeCom 2023

作者： Mei, Ben-Xia Chen, Lin-Shu Sun, Shao-Jie Chen, Pan-Yu Huang, Wei-Liang Hunan University of Science and Technology School of Computer Science and Engineering XiangTan China

ISBN: (纸本)9798350312461

With the problems of handling single-feature and overlooking user preferences in the recommendation algorithms, this paper proposes a Recurrent Neural NetWork-based Multi-feature Hybrid Recommendation Model (RN-MHRM). Firstly, features are extracted from user-item interaction data using the Latent Factor Model (LFM), and an improved Recurrent Neural NetWork (RNN) is used to replace the linear inner product of LFM vectors With non-linearity, Which aims at learning richer features that capture user's short-term interests. Secondly, to avoid single-feature, item information is introduced and the BERT model is used for extracting multi-features. Thirdly, both short-term and long-term interests are considered, and the user's long-term interests are trained by a FeedforWard Neural NetWork (FNN), Which greatly improves the recommendation performance. Experiments designed on multiple real datasets have shown that RN-MHRM effectively improves recommendation performance compared to the baseline model. © 2023 IEEE.

关键词： Recommender systems

来源：评论

学校读者我要写书评

暂无评论

F-CACNN: A framework integrating FNet and adaptive attention for end-to-end EEG emotion recognition 9

F-CACNN: A framework integrating FNet and adaptive attention...

引用

9th IEEE Smart World Congress, SWC 2023

作者： Cao, Zhoulei Wang, Xiaoliang Liu, Yuzhen Hunan University of Science and Technology School of Computer Science and Engineering Xiangtan China

ISBN: (纸本)9798350319804

An important area of study in human-computer interaction is emotional brain-computer interface based on electroencephalogram (EEG). A challenge in this area is how to make full use of channel correlation and provide timely feedback on effective emotion recognition results. In this study, we present a FNet-based and channel attention convolution neural network-based model (F-CACNN) for emotion recognition. We thoroughly explore the EEG channel information with the adaptive channel attention mechanism and employ the FNet mechanism for the first time to realize emotion recognition more efficiently. On the one hand, we include the adaptive channel attention technique into the convolutional neural network to adaptively assign the weights according to the importance of each channel and extract EEG spatial features. On the other hand, the FNet mechanism is integrated into long short-term memory. It takes the Fourier Transform to extract the EEG temporal feature without learnable parameterization. Experimental results show that our method can effectively realize emotion recognition without handcrafted features extraction, and significantly reduce time consumption, which meets the current demand for end-to-end emotion recognition. © 2023 IEEE.

关键词： Affective brain-computer interface Channel attention mechanism EEG Emotion recognition Fourier Transform

来源：评论

学校读者我要写书评

暂无评论

Effect of crystal morphology of ultrahigh-nickel cathode materials on high temperature electrochemical stability of lithium ion batteries

引用

Journal of Energy Chemistry 2024年第1期88卷 327-335,I0008页

作者： Bi Luo Hui Li Haoyu Qi Yun Liu Chuanbo Zheng Weitong Du Jiafeng Zhang Lai Chen School of Metallurgy Engineering Jiangsu University of Science and TechnologyZhangjiagang 215600JiangsuChina National Engineering Laboratory for High-Efficiency Recovery of Refractory Nonferrous Metals School of Metallurgy and EnvironmentCentral South UniversityChangsha 410083HunanChina School of Materials Science and Engineering Beijing Key Laboratory of Environmental Science and EngineeringBeijing Institute of TechnologyBeijing 100081China

Higher nickel content endows Ni-rich cathode materials LiNi_(x)Co_yMn_(1-x-y)O_(2)(x>0.6)with higher specific capacity and high energy density,which is regarded as the most promising cathode materials for Li-ion ***,the deterioration of structural stability hinders its practical application,especially under harsh working conditions such as high-temperature *** these circumstances,it becomes particularly critical to clarify the impact of the crystal morphology on the structure and high-temperature performance as for the ultrahigh-nickel ***,we conducted a comprehensive comparison in terms of microstructure,high-temperature long-cycle phase evolution,and high-temperature electrochemical stability,revealing the differences and the working mechanisms among polycrystalline(PC),single-crystalline(SC)and Al doped SC ultrahigh-nickel *** results show that the PC sample suffers a severe irreversible phase transition along with the appearance of microcracks,resulting a serious decay of both average voltage and the energy *** the Al doped SC sample exhibits superior cycling stability with intact layered ***-situ XRD and intraparticle structural evolution characterization reveal that Al doping can significantly alleviate the irreversible phase transition,thus inhibiting microcracks generation and enabling enhanced ***,it exhibits excellent cycling performance in pouch-type full-cell with a high capacity retention of 91.8%after 500 cycles at 55℃.This work promotes the fundamental understanding on the correlation between the crystalline morphology and high-temperature electrochemical stability and provides a guide for optimization the Ni-rich cathode materials.

关键词： Single-crystalline Ultrahigh-nickel cathode High-temperature performance Phase evolution

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：