检索结果-内蒙古大学图书馆

Knowledge-Guided Semantic Transfer Network for Few-Shot Image Recognition

IEEE TRANSACTIONS ON NEURAL NETWORKS AND LEARNING SYSTEMS 2023年第2期PP卷 3573-3586页

作者： Li, Zechao Tang, Hao Peng, Zhimao Qi, Guo-Jun Tang, Jinhui Nanjing Univ Sci & Technol Sch Comp Sci & Engn Nanjing 210094 Peoples R China Nankai Univ Coll Comp Sci Tianjin 300350 Peoples R China Maple Lab Net Lab MAchine Percept & Learning MAPLE Orlando FL 32816 USA Futurewei Technol Seattle WA 98004 USA

Deep learning-based models have been shown to outperform human beings in many computer vision tasks with massive available labeled training data in learning. However, humans have an amazing ability to easily recognize images of novel categories by browsing only a few examples of these categories. In this case, few-shot learning comes into being to make machines learn from extremely limited labeled examples. One possible reason why human beings can well learn novel concepts quickly and efficiently is that they have sufficient visual and semantic prior knowledge. Toward this end, this work proposes a novel knowledge-guided semantic transfer network (KSTNet) for few-shot image recognition from a supplementary perspective by introducing auxiliary prior knowledge. The proposed network jointly incorporates vision inferring, knowledge transferring, and classifier learning into one unified framework for optimal compatibility. A category-guided visual learning module is developed in which a visual classifier is learned based on the feature extractor along with the cosine similarity and contrastive loss optimization. To fully explore prior knowledge of category correlations, a knowledge transfer network is then developed to propagate knowledge information among all categories to learn the semantic-visual mapping, thus inferring a knowledge-based classifier for novel categories from base categories. Finally, we design an adaptive fusion scheme to infer the desired classifiers by effectively integrating the above knowledge and visual information. Extensive experiments are conducted on two widely used Mini-ImageNet and Tiered-ImageNet benchmarks to validate the effectiveness of KSTNet. Compared with the state of the art, the results show that the proposed method achieves favorable performance with minimal bells and whistles, especially in the case of one-shot learning.

关键词： visualization Semantics Image recognition Task analysis Knowledge transfer Knowledge engineering Feature extraction Few-shot learning image recognition knowledge graph knowledge transfer

来源：评论

学校读者我要写书评

暂无评论

Exploratory and statistical data analysis of Indian economy 5

Exploratory and statistical data analysis of Indian economy

引用

5th IEEE International conference on Advances in Computing, Communication Control and Networking, ICAC3N 2023

作者： Singh, Bhawna vandana Umrao, Sandhya Katiyar, Sanjay Sharma, vanshika G.L. Bajaj Institute Of Technology And Management Department Of Cse Greater Noida India Noida Institute Of Engineering And Technology Department Of Cse Greater Noida India Greater Noida Institute Of Engineering And Technology Department Of Cse Greater Noida India

ISBN: (纸本)9798350330861

Exploratory data analysis provides visual exploration of data and also allows for the identification of trends and relationships between variables. The aim of this research is to study about different factors involved in the economy of India. The dataset has been collected from year 1960 to 2020 and is analyzed by looking at the factors such as GDP, import and export value of goods and services, total reserves(include gold, current US Dollar) and health status of nation's citizens. Thus, it suggests that all factors are important for a stable economy and it also provides different ways to improve these factors. © 2023 IEEE.

关键词： Information analysis

来源：评论

学校读者我要写书评

暂无评论

EEG-Based Emotion Recognition Using Principal Component analysis and Support vector Machine

EEG-Based Emotion Recognition Using Principal Component Anal...

引用

2024 International conference on Control and Automation, Electronics, Robotics, Internet of Things, and Artificial Intelligence, CERIA 2024

作者： Tran, Minh Anh Nguyen, Le Hoa The University of Danang University of Science and Technology Danang Viet Nam

ISBN: (纸本)9798331511166

Recognition of human's emotional states is one of crucial issues in many Brain-Computer-Interface (BCI) applications. In the current study, the problem of emotion recognition based on EEG data will be addressed. The public SEED-v EEG dataset acquired during affect elicitation by audio-visual stimuli was first presented. Then, the preprocessing techniques including noise filtering, down sampling, trail separation were employed on the raw data. To reduce the size of data as well as remove less valuable data, the Principal Component analysis (PCA) was proposed. Next, the Power Spectral Density (PSD) method was used to extract emotional features. Finally, we utilized the Support vector Machine (SvM) for emotion classification. The obtained results show that a high classification accuracy (up to 95.6%) can be achieved. In addition, the obtained results also demonstrate that the proposed PCA technique greatly contributes to improving classification accuracy. © 2024 IEEE.

关键词： electroencephalogram emotional state power spectral density principal component analysis support vector machine

来源：评论

学校读者我要写书评

暂无评论

Highly Efficient HDR Signal Processing for ASICs Using the Flexfloat data Type 26

Highly Efficient HDR Signal Processing for ASICs Using the F...

引用

26th International conference on Digital Signal Processing and its Applications, DSPA 2024

作者： Yu, Zinovchik O. Chesnokov, v.N. Zaripov, R.A. Moscow Institute of Physics and Technology Russia

ISBN: (纸本)9798350361728

The on-going evolution of technology of capturing and generation video, sound, RF, etc. signals results in very HDR data to be stored and processed. The analysis of human perception of different signals is showing that the linear representation of the above mentioned signals in ASICs is highly inefficient (has redundant bits corresponding to high values of the signal). This results in excessive power consumption and silicon area. This paper is presenting a new datatype (FlexFloat), which is designed taking into account the behavior of Human visual System and allows highly efficient processing and storage for HDR data. The effectiveness of the proposed methodology is demonstrated on the example of HDR image processing. © 2024 IEEE.

关键词： Field programmable gate arrays (FPGA)

来源：评论

学校读者我要写书评

暂无评论

Context-Based Annotation visualisation in virtual Reality: A Use Case in Archaeological data exploration 21st

Context-Based Annotation Visualisation in Virtual Reality: A...

引用

21st EuroXR International conference, EuroXR 2024

作者： De Bonis, Michele Nguyen, Huyen Bourdot, Patrick Université Paris-Saclay CNRS LISN Gif-sur-Yvette France

ISBN: (纸本)9783031785924

In this paper we present a virtual Reality (vR) tool that facilitates the visualisation and exploration of context-based, multi-level annotations in archaeology. Indeed, thanks to photogrammetry and laser scanning techniques, archaeologists can capture and reconstruct almost faithfully ruins and excavations in 3D. Thus, they can maximise their work performance on the field within their limited excavation time and leave deep, but often slow, analysis tasks at later stages offsite. Using vR, we can enrich 3D captured and reconstructed environments by adding extra information for ex situ analysis of archaeological data. This approach can also help the archaeologists speed up their collaboration with other experts or easily share the information with the public. We aim to assist the exploration of numerous annotations and information at disposal that can be densely clustered in a 3D reconstructed world. We collaborated closely with archaeologists at AOROC in the process of designing annotations that convey information with increasing Levels of Detail (LoDs), reflecting the complexity of the contained data. We then developed different interaction techniques that allow users to explore the annotations by switching between LoDs, including active selection by clicking, proximity approach, and the combination of both. Additionally, we proposed and evaluated the effectiveness of annotation grouping mechanism to enhance visual clarity by aggregating annotations in close spatial proximity. The results of our user study with non-experts, aggregated with a qualitative study with the archaeologists, showed user preference for the combination of the interaction techniques, but evidenced some limitations for the grouping of annotations. © The Author(s), under exclusive license to Springer Nature Switzerland AG 2025.

关键词： virtual reality

来源：评论

学校读者我要写书评

暂无评论

Emotion Recognition from RGB Kinect videos and Physiological Signals Using Pre-Trained CNN Model 3

Emotion Recognition from RGB Kinect Videos and Physiological...

引用

3rd International conference on Intelligent Computing and Next Generation Networks, ICNGN 2024

作者： Anand, Shruthi Mookiah, Lenin Bakkialakshmi, v.S. Surakshitha, S. Sembon Department of Computing Technologies SRM Institute of Science and Technology Kattankalathur India Nevada United States

ISBN: (纸本)9798331529222

This study presents an innovative approach to emotion recognition by integrating RGB Kinect video data with physiological signals, including electroencephalography (EEG), electrocardiography (ECG), and galvanic skin response (GSR). Using the AMIGOS dataset, which includes video recordings and self-assessment emotional ratings, we improve emotion prediction through a comprehensive multimodal analysis. Our method involves feature extraction and averaging from both video and physiological signals, emphasizing the connection between visual cues and emotional responses. A pre-trained ResNet model is used for advanced feature extraction from video data. various machine learning algorithms are applied to evaluate predictive models against self-assessment labels. Results indicate that the combination of ResNet-based visual features and physiological signals enhances emotion prediction accuracy, underscoring the value of a multimodal approach in capturing complex emotional dynamics. ©2024 IEEE.

关键词： Physiological models

来源：评论

学校读者我要写书评

暂无评论

Interactive visualization of BioMedical data

Interactive Visualization of BioMedical Data

引用

2024 IEEE International conference on Big data, Bigdata 2024

作者： Islam, Mohammad Shafiqul Ibn Rafiq, Rahat Tusch, Guenter Grand Valley State University College of Computing MI United States

ISBN: (纸本)9798350362480

Public databases such as the NCBI Gene Expression Omnibus (GEO) house millions of experimental gene expression datasets invaluable for transcriptome meta-analysis, enabling researchers to identify genes, pathways, and biomarkers with enhanced accuracy. However, while these datasets offer great potential, finding relevant biological pathways in disease-related contexts remains challenging. To address this, we introduce a web-based visualization tool that utilizes PubMed abstracts to extract key biomedical entities, such as genes, pathways, diseases, species, and chemicals, and represent their relationships through an interactive graph. By leveraging PubMed's annotated MeSH terms, our tool allows users to explore the literature more intuitively, bypassing traditional dataset search limitations and focusing on published results. The tool transforms complex query outputs into a graph-based network, providing a visual interface for navigating connections between biological entities. This interactive approach helps researchers to analyze relationships within large-scale biomedical data and uncover potential pathways for drug repurposing and disease understanding. We discuss the tool's design using ***, its application in biomedical research, and the challenges faced when visualizing large datasets, offering new insights for improved data interpretation and exploration. © 2024 IEEE.

关键词： BioInformatics BioMedical data visualization PubMed

来源：评论

学校读者我要写书评

暂无评论

visual Analytics for Profiling Land Use Changes 36

Visual Analytics for Profiling Land Use Changes

引用

36th SIBGRAPI conference on Graphics, Patterns and Images (SIBGRAPI)

作者： Santos, Claudio Hosseini, Maryam Rulff, Joao Miranda, Fabio Wilson, Luc Silva, Claudio Ferreira, Nivan Lage, Marcos Univ Fed Fluminense Niteroi RJ Brazil MIT Cambridge MA USA NYU New York NY USA Univ Illinois Chicago IL USA Kohn Pedersen Fox Associates PC New York NY USA Univ Fed Pernambuco Recife Brazil

ISBN: (纸本)9798350338737;9798350338720

The growth of cities calls for regulations and zoning rules on how each piece of urban space will be used. Tracking land use can reveal a wealth of information about urban development. For that matter, cities have been releasing data sets describing the historical evolution of the shape and the attributes of land units. The complex nature of land-use data, however, makes the analysis of such data challenging and time-consuming. To address these challenges, we propose URBAN CHRONICLES, a visual analytics system that enables interactive exploration of land-use changes. Using New York City's Primary Land Use Tax Lot Output (PLUTO), we show the system's capabilities to explore the data from several years at different scales. URBAN CHRONICLES supports on-the-fly aggregation and filtering operations that leverage the hierarchical nature of the data set to index the shape and attributes of geographical regions that change over time. Finally, we demonstrate the system's utility through case studies that analyze the impact of Hurricane Sandy on land use attributes and the effects of rezoning plans in Brooklyn.

关键词： visual analytics

来源：评论

学校读者我要写书评

暂无评论

Revolutionizing leather quality assurance through deep learning powered precision in defect detection and segmentation by a comparative analysis of Mask RCNN and YOLO v8

Revolutionizing leather quality assurance through deep learn...

引用

2024 International conference on Advances in data Engineering and Intelligent Computing Systems, ADICS 2024

作者： Prathyuksha Nair, J. Thangakumar, J. Hindustan University Department Of Computer Science And Engineering Chennai India

ISBN: (纸本)9798350364828

The leather industry confronts persistent challenges in the maintenance of its product quality, often produced with defects that escape from the traditional conventional inspection methods. This research seeks the transformation of leather quality assurance by harnessing the power of deep learning, specifically through the comparative analysis exploration of Mask RCNN (Mask Region Convolutional Neural Network) and YOLO v8 (You Only Look Once version 8) for precise defect detection and segmentation of wet blue leather. In this comprehensive analysis, the performance of Mask RCNN and YOLO v8 are evaluated using an industrial dataset comprising various defect categories. The study encompasses a deep dive into the intricacies of each model's architecture, training processes, and inference speed to provide an obvious understanding of their respective limitations and strength. Findings from the research showcase the efficacy of both Mask RCNN and YOLO v8 in significantly enhancing defect detection accuracy compared to traditional methods. However, distinct differences emerge in terms of computational speed, resource requirements, and the models' ability to handle diverse defect sizes and shapes. This research provides valuable insights for industry practitioners, being in the selection of the most suitable model for specific operational contexts and facilitates the more informed decisions in implementation of deep learning-powered quality assurance systems. By advancing the precision of defect detection and segmentation through deep learning, the experiment results provided an inference of YOLOv8 being the model with more accuracy in detection and segmenting the defects in the leather, also its ability to provide and predict the defects using lesser training data and time, makes it more efficient model for the manufacturing of good quality of leather. © 2024 IEEE.

关键词： Computer vision

来源：评论

学校读者我要写书评

暂无评论

Navigating Exascale Operational data Analytics: From Inundation to Insight

Navigating Exascale Operational Data Analytics: From Inundat...

引用

2024 Workshops of the International conference for High Performance Computing, Networking, Storage and analysis, SC Workshops 2024

作者： Shin, Woong Osborne, Tim Karimi, Ahmad Maroof Palumbo, Rachel May, Alex Lester, Corwin Hines, Jesse Sattar, Naw Safrin Huk, Leah Simmerman, Scott Brewer, Wesley Miller, Jeffrey Adamson, Ryan Kuchar, Olga Prout, Ryan Wang, Feiyi Atchley, Scott Oral, Sarp Oak Ridge National Laboratory Oak RidgeTN United States

ISBN: (纸本)9798350355543

In this paper, we address the challenges in achieving sustainable data-driven efficiency by providing a detailed exploration of the end-to-end operational data analytics (ODA) framework that evolved through two generations of supercomputer systems at the Oak Ridge Leadership Computing Facility (OLCF). This framework addresses large data streams ingested from heavily instrumented HPC environment that accumulates multi-terabytes per day. We outline the multifaceted data life cycle across HPC procurement, operations, and research & development, identifying key obstacles and design decisions that shape effective strategies in building and supporting data pipelines end-to-end. By sharing key insights and lessons learned from our experience, we offer recommendations for the HPC community on enabling sustainable operational data analytics and beyond. Our contributions aim to bridge the gap between potential and real benefits of operational data, guiding future efforts towards integrated and sustainable operational intelligence in high-performance computing environments. © 2024 IEEE.

关键词： data Analytics data Governance HPC Post-Exascale Challenges Machine Learning Applications Monitoring Operational data Analytics Telemetry visual Analytics

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：