检索结果-内蒙古大学图书馆

Self-Supervised Representation Learning: Introduction, advances, and challenges

ieee signal processing MAGAZINE 2022年第3期39卷 42-62页

作者： Ericsson, Linus Gouk, Henry Loy, Chen Change Hospedales, Timothy M. Univ Edinburgh Edinburgh EH8 9AB Midlothian Scotland Univ Waikato Hamilton New Zealand Univ Edinburgh Sch Informat Edinburgh EH8 9AB Midlothian Scotland Nanyang Technol Univ Singapore 639798 Singapore Nanyang Technol Univ MMLab NTU Singapore 639798 Singapore Chinese Univ Hong Kong Hong Kong Peoples R China Univ Edinburgh Artificial Intelligence Edinburgh EH8 9AB Midlothian Scotland Samsung AI Ctr Cambridge Cambridge CB1 2JH England Samsung AI Ctr Cambridge Machine Learning & Data Intelligence Programme Cambridge CB1 2JH England

Self-supervised representation learning (SSRL) methods aim to provide powerful, deep feature learning without the requirement of large annotated data sets, thus alleviating the annotation bottleneck-one of the main barriers to the practical deployment of deep learning today. These techniques have advanced rapidly in recent years, with their efficacy approaching and sometimes surpassing fully supervised pretraining alternatives across a variety of data modalities, including image, video, sound, text, and graphs. This article introduces this vibrant area, including key concepts, the four main families of approaches and associated state-of-the-art techniques, and how self-supervised methods are applied to diverse modalities of data. We further discuss practical considerations including workflows, representation transferability, and computational cost. Finally, we survey major open challenges in the field, that provide fertile ground for future work.

关键词： Representation learning Deep learning Annotations computational efficiency

来源：评论

学校读者我要写书评

暂无评论

CMCL: Cross-Modal Compressive Learning for Resource-Constrained Intelligent IoT Systems

引用

ieee INTERNET OF THINGS JOURNAL 2024年第15期11卷 25534-25542页

作者： Chen, Bin Tang, Dong Huang, Yujun An, Baoyi Wang, Yaowei Wang, Xuan Harbin Inst Technol Shenzhen Sch Comp Sci & Technol Shenzhen 518055 Guangdong Peoples R China Guangdong Prov Key Lab Novel Secur Intelligence Te Shenzhen Peoples R China Peng Cheng Lab Res Ctr Artificial Intelligence Shenzhen 518055 Guangdong Peoples R China Tsinghua Univ Tsinghua Shenzhen Int Grad Sch Shenzhen 518055 Guangdong Peoples R China Huawei Technol Co Ltd Network Technol Lab Shenzhen 518055 Peoples R China

Compressive learning (CL) has proven to be highly successful in executing joint signal sampling and inference for intricate vision tasks through resource-limited Internet of Things (IoT) devices. Recent studies have turned their attention toward utilizing the deep neural networks (DNNs) methodology, also known as DeepCL, to enhance performance in unimodal vision tasks. This approach incorporates learnable compressed sensing in a comprehensive, end-to-end manner. Current DeepCL techniques typically employ initial signal reconstruction as the input for subsequent DNNs for inference. However, this practice presents potential risks, such as privacy breaches and reduced performance due to information processing inequality. To address these issues, this article introduces the first cross-modal CL (CMCL) approach that enables image captioning directly on compressed measurements. When compared to previous DeepCL strategies, the proposed CMCL offers significant improvements in computational efficiency and privacy protection. Extensive experiments demonstrate that CMCL performance is nearly on par with leading image captioning methods, showcasing a metric value that is merely 2.75% lower than the uncompressed method when the data is compressed eightfold.

关键词： Compressed sensing Task analysis Internet of Things image coding Transformers Servers Privacy Compressive learning (CL) image captioning Internet of Things (IoT) transformer

来源：评论

学校读者我要写书评

暂无评论

Thresholds decision method for fast object detection systems

Thresholds decision method for fast object detection systems

引用

ieee International symposium on computational intelligence in Robotics and Automation

作者： Sato, H Mitsukura, Y Fukumi, M Akamatsu, N Aoyama Gakuin Univ Sagamihara Kanagawa 2298558 Japan

ISBN: (纸本)0780378660

In this paper, we propose a new determination method of color thresholds for extracting objects. In order to obtain the optimal color threshold, it is necessary to determine RGB values all together because one color consists of a combination of RGB values. The two optimal thresholds, a upper and a lower, axe determined for object extraction by applying genetic algorithm. Moreover, in the proposed method, individuals axe evaluated using two fitness functions for obtaining the optimal color thresholds, and initial population is generated with color histograms for fast convergence. Finally, computer simulations axe performed, and the result images are shown.

关键词： image processing object extraction threshold genetic algorithm fitness function

来源：评论

学校读者我要写书评

暂无评论

A smart sensor for image processing: Towards a system on chip

A smart sensor for image processing: Towards a system on chi...

引用

ieee International symposium on Industrial Electronics

作者： Elouardi, A. Bouaziz, S. Dupret, A. Lacassagne, L. Klein, J. O. Reynaud, R. Univ Paris 11 Inst Fundamental Elect Bat 220 F-91405 Orsay France

ISBN: (纸本)9781424404964

One of the solutions to reduce the computational complexity of image processing is to perform some low-level computations on the sensor focal plane. This paper presents a vision system based on a smart sensor. PARIS1. (Programmable Analog Retina-like image Sensor) is the first prototype used to evaluate the architecture of an on-chip vision system based on such a sensor and a digital processor. The sensor integrates analog and digital computing units. This architecture makes the vision system more compact and increases the performances reducing the data flow exchanges with the digital processor. A system has been implemented as a proof-of-concept. This has enabled us to evaluate the performance needed for a possible implementation of a digital processor on the same chip. The approach is compared to two architectures implementing CMOS sensors and interfaced to the same processor. The comparison is related to image processing computation time, processing reliability, programmability, precision, bandwidth and subsequent stages of computations.

关键词： CMOS image sensors computational complexity digital signal processing chips focal planes image processing intelligent sensors system-on-chip CMOS sensors PARISl data flow digital processor on-chip vision system p

来源：评论

学校读者我要写书评

暂无评论

Text generation by probabilistic suffix tree language model 16

Text generation by probabilistic suffix tree language model

引用

16th International Joint symposium on Artificial intelligence and Natural Language processing (ISAI-NLP)

作者： Marukatat, Sanparith NECTEC Image Proc & Understanding Team AI Res Grp Bangkok Thailand

ISBN: (纸本)9781665409476

During last decade, language modeling has been dominated by neural structures;RNN, LSTM or Transformer. These neural language models provide excellent performance to the detriment of very high computational cost. This work investigates the use of probabilistic language model that requires much less computational cost. In particular, we are interested in variable-order Markov model that can be efficiently implemented on a probabilistic suffix tree (PST) structure. The PST construction is cheap and can be easily scaled to very large dataset. Experimental results show that this model can be used to generated realistic sentences.

关键词： Thai NLP language model probabilistic suffix tree

来源：评论

学校读者我要写书评

暂无评论

computational experiments with area-based stereo for image-based rendering 1

Computational experiments with area-based stereo for image-b...

引用

1st International symposium on 3D Data processing Visualization and Transmission

作者： Izquierdo, E Univ London Queen Mary Coll Dept Elect Engn Multimedia & Vis Lab London E1 4NS England

ISBN: (纸本)0769515223

In this paper two disparity estimators with different complexity degrees are described and used to examine how much disparity inaccuracies influence image rendering quality. The objective of this study is to design software-based image synthesis in real-time on conventional PC platforms. Basically, this work looks at the opposite end of the cost-complexity curve by making very restrained demands to the disparity estimator. It is empirically shown that in many cases the effect of disparity accuracy in the quality of virtual views is almost imperceptible and that for many applications requiring real-time processing reasonable good results can be achieved with less computational cost.

关键词： Stereo image processing

来源：评论

学校读者我要写书评

暂无评论

An improved cartoon+texture decomposition based pansharpening method 19

An improved cartoon+texture decomposition based pansharpenin...

引用

19th CSI International symposium on Artificial intelligence and signal processing, AISP 2017

作者： Lotfi, Meysam Ghassemian, Hassan Image Processing and Information Analysis Lab. Faculty of Electrical and Computer Engineering Tarbiat Modares University Tehran Iran

ISBN: (纸本)9781538625859

Pansharpening is the most widely used fusion method, in the field of remote sensing, to increase spatial information of the multispectral image while preserving spectral signatures. Based on the nature of spatial and spectral information, there is a lack of correlation between them. Therefore, separation of them can be considered as an image decomposition to uncorrelated components. Recently, the cartoon+texture decomposition was used in the pansharpening and decrease spectral distortion. However, details have not been strengthened enough. Therefore, in this paper we aim to use a filter based detail extraction to improve spatial information while mitigate spectral distortion. © 2017 ieee.

关键词： Low pass filters

来源：评论

学校读者我要写书评

暂无评论

Research on method in detecting cavity of the cross-section of fiberboard based on image processing

Research on method in detecting cavity of the cross-section ...

引用

International symposium on intelligence Information processing and Trusted Computing

作者： Zhang, Jian-Fei Ma, Yan Forestry and Wood Working Mechanical Engineering Technology Center Northeast Forestry University Harbin China College of Computer and Information Engineering Heilongjiang Institute of Science and Technology Harbin China

ISBN: (纸本)9780769541969

Aiming at the feature of the cavity microscopic image of the cross-section of Fiberboard, a new algorithm is proposed which can detect the image effectively and accurately. This algorithm used Otsu algorithm to segment the cavity from the microscopic image, then refined the cavity edge by using the GAP statistic model which was based on the difference in distribution function of grey-level . The experimental results show that the algorithm improves the edge detection effect of cavity microscopic image with a good anti-noise performance, and achieves better detection effect. © 2010 ieee.

关键词： image segmentation

来源：评论

学校读者我要写书评

暂无评论

Two Methods of Determining Target Orientation by Robot Visual Principle

Two Methods of Determining Target Orientation by Robot Visua...

引用

10th International symposium on computational intelligence and Design (ISCID)

作者： Liu, Gang Tianjin Univ Technol Sch Mech Engn Tianjin Key Lab Adv Mech & Elect Syst Design & In Tianjin 300384 Peoples R China

ISBN: (纸本)9781538636749

by analyzing the imaging principle of the fish-eye lens in the robot vision, two methods of determination of the target orientation based on the fish-eye lens are summarized. One method uses the distortion correction to determine the target orientation, and the other uses the relationship between the target in the actual orientation and its correspondence in the imaging coordinates to determine the target orientation, both methods all use image processing technology, fast and efficiently determine the orientation of the target in the fish eye image. In this paper, the theoretical basis and experimental process of the two methods are introduced respectively. By comparing the data after the experiment, the more accurate method is analyzed and applied to the target azimuth determination of the fish-eye lens.

关键词： fish-eye lens robot vision target orientation image processing

来源：评论

学校读者我要写书评

暂无评论

A framework for initialising a dynamic clustering algorithm: ART2-A 5

A framework for initialising a dynamic clustering algorithm:...

引用

ieee symposium on computational intelligence and Data Mining (CIDM)

作者： Chambers, Simon J. Jarman, Ian H. Lisboa, Paulo J. G. Liverpool John Moores Univ Sch Comp & Math Sci Liverpool L3 3AF Merseyside England

ISBN: (纸本)9781479945184

Algorithms in the Adaptive Resonance Theory (ART) family adapt to structural changes in data as new information presents, making it an exciting candidate for dynamic online clustering of big health data. Its use however has largely been restricted to the signal processing field. In this paper we introduce an refinement of the ART2-A method within an adapted separation and concordance (SeCo) framework which has been shown to identify stable and reproducible solutions from repeated initialisations that also provides evidence for an appropriate number of initial clusters that best calibrates the algorithm with the data presented. The results show stable, reproducible solutions for a mix of real-world heath related datasets and well known benchmark datasets, selecting solutions which better represent the underlying structure of the data than using a single measure of separation. The scalability of the method and it's facility for dynamic online clustering makes it suitable for finding structure in big data.

关键词： Big Data adaptive resonance theory medical information systems pattern clustering ART2-A SeCo adapted separation and concordance framework adaptive resonance theory family benchmark datasets big health data dynamic clustering algorithm dynamic online clustering real-world heath related datasets signal processing field Big data Breast cancer Clustering algorithms Heuristic algorithms Prototypes Subspace constraints Vectors big data Clustering algorithms adaptive resonance theory medical information systems Heuristic algorithms Cluster Analysis Subspace constraints signal processing PROTOTYPE Breast Cancer

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：