检索结果-内蒙古大学图书馆

Conference on computer Vision and Pattern Recognition (CVPR)

作者： Beichen Zhang Xiaoxing Wang Xiaohan Qin Junchi Yan Department of Computer Science and Engineering & MoE Key Lab of AI Shanghai Jiao Tong University

ISBN: (数字)9798350353006

ISBN: (纸本)9798350353013

Supernet is a core component in many recent Neural Architecture Search (NAS) methods. It not only helps embody the search space but also provides a (relative) estimation of the final performance of candidate architectures. Thus, it is critical that the top architectures ranked by a supernet should be consistent with those ranked by true performance, which is known as the order-preserving ability. In this work, we analyze the order-preserving ability on the whole search space (global) and a sub-space of top architectures (local), and empirically show that the local order-preserving for current two-stage NAS methods still need to be improved. To rectify this, we propose a novel concept of Supernet Shifting, a refined search strategy combining architecture searching with supernet fine-tuning. Specifically, apart from evaluating, the training loss is also accumulated in searching and the supernet is updated every iteration. Since superior architectures are sampled more frequently in evolutionary searching, the supernet is encouraged to focus on top architectures, thus improving local order-preserving. Besides, a pre-trained supernet is often un-reusable for one-shot methods. We show that Supernet Shifting can fulfill transferring supernet to a new dataset. Specifically, the last classifier layer will be unset and trained through evolutionary searching. Comprehensive experiments show that our method has better order-preserving ability and can find a dominating architecture. Moreover, the pre-trained supernet can be easily transferred into a new dataset with no loss of performance.

关键词： Training computer vision Estimation computer architecture Search problems Boosting Pattern recognition

来源：评论

学校读者我要写书评

暂无评论

Cutting-Edge Travel Planner: Intelligent Route Recommendation System using Enhanced Learning Scheme with ai Principles 9

Cutting-Edge Travel Planner: Intelligent Route Recommendatio...

引用

9th International Conference on science, Technology, engineering and Mathematics, ICONSTEM 2024

作者： Gladiss Merlin, N.R. Anandakumar, D. Devilavanya, S. Kumari, P.K. Sheela Shantha Jayakumar, T. R.M.K Engineering College Department of Ai&ds Chennai Kavarapettai India Nandha Engineering College Department of Ece Erode India Dr. Mahalingam College of Engineering And Technology Department of Computer Science And Engineering Pollachi India VeltechHightech Dr Rangarajan Dr Sakunthala Engineering College Department of Cse Chennai Avadi India Kcg College of Technology Department of Mechatronics Engineering Chennai India

ISBN: (纸本)9798350365092

A route planner is crucial as it optimizes travel efficiency, minimizes time and fuel consumption, and enhances overall navigation convenience and safety. This paper presents the design and implementation of CollabRouteNet, an intelligent route recommendation system that leverages collaborative filtering and reinforcement learning principles. The proposed model aims to provide personalized and contextually relevant route recommendations tailored to individual user preferences and dynamic environmental conditions. Collaborative filtering techniques are employed to analyze user interactions and discern patterns within historical route data. This involves constructing a user-item matrix and applying matrix factorization to learn latent representations for users and routes. In parallel, reinforcement learning is utilized to optimize route recommendations in real-time by defining the problem as a Markov Decision Process (MDP) and training an agent to learn an optimal route selection policy. The model balances exploration and exploitation to adaptively recommend routes that optimize user satisfaction and navigation efficiency. Implemented in PyCharm, the CollabRouteNet model demonstrates promising results in providing accurate and responsive route recommendations. Through its integrated approach, CollabRouteNet offers a promising solution for enhancing navigation experiences in diverse urban and rural settings. © 2024 IEEE.

关键词： Reinforcement learning

来源：评论

学校读者我要写书评

暂无评论

Decoding Sentiments: An Efficient Trans-Long Short Term Memory to Understand Viewer Reactions on Ad-Supported Video Contents 3

Decoding Sentiments: An Efficient Trans-Long Short Term Memo...

引用

3rd IEEE International Conference on Blockchain and Distributed Systems Security, ICBDS 2024

作者： Ramagundam, Shashishekhar Karne, Niharika Comcast Software Development & Engineering in Content Discovery & Ai Department of Computer Science United States Troy University Montgomery Campus AL United States Dt Amtrak C & Rs Platform Services Department of Electrical Engineering United States University of South Alabama United States

ISBN: (纸本)9798350354348

Viewer reaction analysis in ad-supported video content focuses on the use of advanced techniques to understand and interpret the emotional responses expressed by viewers during ad viewing experiences. This analysis goes beyond simple positive or negative sentiment and delves into the nuances of viewer reactions, allowing for a more comprehensive understanding of the emotional impact of ads. With this information, advertisers can make data-driven decisions to optimize their advertising strategies, create more engaging content, and ultimately enhance the viewer experience. One of the challenges in sentiment analysis for viewer reaction analysis is the inherent subjectivity and variability of human emotions. Emotions can be complex and subjective, making it difficult to accurately capture and interpret them solely based on text or viewer feedback. Different individuals may perceive and express emotions differently, adding another layer of complexity to the analysis process. Addressing these challenges requires a combination of advanced deep learning model refinement. Overcoming these challenges can provide valuable insights into viewer reactions, enabling advertisers and content creators to create more engaging and impactful ad-supported video content. In this paper, an Artificial Intelligence (ai)-based sentiment analysis is introduced to analyze the viewer's reaction while watching the ad in the video content. Here, an efficient model named Trans-Long Short-Term Memory (Trans-LSTM) is developed to perform the sentiment analysis. The developed Trans-LSTM is helpful in supervising the emotional data in the online video content and identifying the popularity of the corresponding video. The suggested model effectively analyzes the reviewer's actions while watching all types of video content. Finally, the experimental analysis is performed to find the effectiveness of the developed sentiment analysis model via various metrics. © 2024 IEEE.

关键词： Video analysis

来源：评论

学校读者我要写书评

暂无评论

An Edge Computing System for Fast Image Recognition Based on Convolutional Neural Network and Petri Net Model

引用

Multimedia Tools and Applications 2024年第5期83卷 12849-12873页

作者： Yang, Cheng-Ying Lin, Yi-Nan Wang, Sheng-Kuan Shen, Victor R.L. Tung, Yi-Chih Lin, Jia-Fu Department of Computer Science University of Taipei 1 Ai-Kao W. Rd. Taipei 100 Taipei Taiwan Department of Electronic Engineering Ming Chi University of Technology 84 Gongzhuan Rd. Taishan District New Taipei City 243 New Taipei Taiwan Department of Electrical Engineering Ming Chi University of Technology 84 Gongzhuan Rd. Taishan District New Taipei City 243 New Taipei Taiwan Department of Computer Science and Information Engineering National Taipei University 151 University Rd. Sanxia District New Taipei City 237 New Taipei Taiwan Department of Information Management Chaoyang University of Technology Taichung Taiwan

As a computer system can precisely detect some target objects, many application scenarios will be developed. In this study, the customized object recognition model was trained by using a software tool, You Only Look Once (YOLO), and burned into one hardware component, Maix Bit, for an image edge computing system, so as to fast recognize humans or objects in the images. In the beginning of system development, a Petri net (PN) model was established to detect all possible abnormal processes and to verify the feasibility and completeness of an edge computing system by using Petri net software tool, Workflow Petri net Designer (WoPeD). Compared with many other development boards, Maix Bit is smaller, more flexible, more economical, and more adaptable to those needs of the edge side scenarios. When equipped with an Artificial Intelligence (ai) chip, Kendryte K210, it is suitable for object recognition due to its superior power consumption, film frames, and processors. YOLO, mainly characterized by its fast speed in using the convolutional neural network (CNN) to recognize objects, can be used to predict the positions and types of multiple objects concurrently. Also, CNN can be used to detect and recognize targets from end to end promptly, when configured to train an image recognition model. Finally, the experimental results have shown that the promising mean Average Precision (mAP), 78.6%, was obtained, which outperforms other existing systems. © 2023, The Author(s), under exclusive licence to Springer science+Business Media, LLC, part of Springer Nature.

关键词： Object detection

来源：评论

学校读者我要写书评

暂无评论

引用

International Conference on Information and Communication Technology Convergence (ICTC)

作者： Gyeong-hyeon Kim Nayoung Ko Eunwoo Kim School of Computer Science and Engineering Chung-Ang University Department of AI Chung-Ang University

ISBN: (数字)9798350364637

ISBN: (纸本)9798350364644

Class-incremental semantic segmentation faces challenges in retaining the knowledge of old classes while learning new ones. Catastrophic forgetting and background shift are the primary hurdles causing performance degradation. Moreover, class confusion arises when new classes share categorical similarities with existing ones. Due to class confusion, the class-incremental semantic segmentation model overwrites the knowledge of old classes with similar new classes. In this paper, we introduce Similarity-aware Class Discrimination (SCD) to address the class confusion problem. SCD increases the distance between classes that share categorical similarities (i.e., visually similar but different classes) in the embedding space, preventing new knowledge from overwriting old knowledge. This approach encourages the model to discriminate between similar categorical classes, mitigating class confusion in class-incremental semantic segmentation. Our method surpasses existing CNN-based and transformer-based class-incremental semantic segmentation methods in PASCAL VOC 2012.

关键词： Degradation Semantic segmentation Semantics Transformers Information and communication technology Faces Convergence

来源：评论

学校读者我要写书评

暂无评论

Enhancing Personalization in Media Streaming Services Through Graph Neural Network 16

Enhancing Personalization in Media Streaming Services Throug...

引用

16th IEEE International Conference on Computational Intelligence and Communication Networks, CICN 2024

作者： Changala, Ravindra Sneha Sri, S. Sandya Kadyan, Sunil Sasthi Kumar, C. Vuyyuru, Veera Ankalu Bala, B Kiran Guru Nanak Institutions Technical Campus Department of Computer Science and Engineering Hyderabad India Krishna University Department of Cse Andhra Pradesh India School of Management and Commerce Manav Rachna University Haryana India Karpagam Academy of Higher Education Department of Computer Science Coimbatore India Koneru Lakshmaiah Education Foundation Department of Cse Andhra Pradesh Vaddeswaram India K.Ramakrishnan College of Engineering Department of Ai & Ds Trichy India

ISBN: (纸本)9798331505264

The increasing diversity of user tastes in media services is a problem for traditional algorithms that aim to provide accurate and customised content choices. This research presents a new method for using Graph Neural Networks (GNNs) to improve personalisation in media streaming services. The links between people, material, and their interactions are sometimes difficult for traditional recommendation algorithms to grasp. GNNs, on the other hand, are excellent at simulating intricate connections in graph-structured data. The suggested approach uses GNNs to build an interaction graph between users and the content, capturing the complex relationships between users, viewing patterns, and content attributes. Through the use of GNNs, people are able to effectively train latent representations that contain client preferences and content attributes in a dynamic manner, leading to suggestions that are more precise and tailored. The suggestion accuracy will continuously increase due to the model's capacity to adapt to changing user behaviour. Extensive trials on real-world video streaming datasets validate the effectiveness of the system, exhibiting better performance than traditional techniques. The Bayesian Graph model lags behind with a greater mean square error (MSE) of 1.25, whilst the GNN shows a competitive MSE of 95, showing effective prediction accuracy for user ratings. © 2024 IEEE.

关键词： Media streaming

来源：评论

学校读者我要写书评

暂无评论

Advancing Plant Disease Identification using Convolutional Neural Networks for Precise Diagnosis in Hibiscus Plants

Advancing Plant Disease Identification using Convolutional N...

引用

Artificial Intelligence and Emerging Technology (Global ai Summit), International Conference on

作者： Nandita Goyal Shivani Agarwal Tripti Choudhary Usikela Naresh Laith H. Alzubaidi Ram Sharma Department of Technology Technology Ajay Kumar Garg Engineering College Ghaziabad India Department of CSE-AIML Hi-Tech Engineering College Ghaziabad India Department of Computer Science and Engineering (AI&ML) CVR College of Engineering Hyderabad India The Islamic University Najaf Iraq Department of Computer Science and Engineering Raj Kumar Goel Institute of Technology Ghaziabad India

ISBN: (数字)9798350379716

ISBN: (纸本)9798350379723

The effectiveness and accuracy of deep learning models for the precise diagnosis of diseases in Hibiscus plants are studied. Gray mold, mainly caused by Botrytis cinerea, in diseased Hibiscus leaves is identified using 560 images collected to form the dataset. Preprocessing of the obtained dataset is performed through resizing, data augmentation, and noise reduction, which ensures the variability of the dataset. The preprocessed dataset is employed to train four different deep learning architectures: custom CNN, ResNet-50, VGG-16, and VGG-19. The benefits of using each model are that they are able to extract features and learn the pattern of the distribution of these features in the presence of the disease. The results are obtained based on the accuracy, precision, recall, F1 score, and the area under the receiver operating characteristic curve for comparison purposes. According to the results, the best candidate is VGG-19, which demonstrates accuracy at the level of 98.54%, while the precision of 98.80%, recall of 98.50%, F1 score of 98.65%, and AUC-ROC of 99.00% are also high. The confusion matrices provide details from which the correct classification for VGG-19 of healthy and diseased leaves becomes evident. The real-time implementation of VGG-19 is completely reliable. The applied deep learning models provide numerous benefits in predicting the diseases in plants, and the usage of computational techniques will ensure food security and opportunities for current development.

关键词： Deep learning Training Plant diseases Accuracy Computational modeling Receivers Feature extraction Convolutional neural networks Reliability Artificial intelligence

来源：评论

学校读者我要写书评

暂无评论

Artificial Intelligence Scoring Algorithm for English Translation Based on DNN 2

Artificial Intelligence Scoring Algorithm for English Transl...

引用

2nd International Conference on Smart Technologies for Smart Nation, SmartTechCon 2023

作者： Priya, R. Krishna Aanandha Saravanan, K. Gadgil, Aashish A. Sabarmathi, K.R. Niranjanamurthy, M. Azarudeen, K. Department of Engineering-Electrical section College of Engineering and Technology University of Technology and Applied Sciences Shinas Oman Department of ECE Veltech Rangarajan Dr.Sagunthala R&D Institute of Science and technology Chennai India Department of ECE KLS GOGTE institute of Technology Karnataka Belagavi India Department of Information Technology Bannari Amman Institute of Technology Tamilnadu Sathyamangalam India Department of AI and ML BMS Institute of Technology and Management Avalahalli. Karnataka Bangalore India Department of Computer Science and Engineering Velammal College of Engineering and Technology Madurai India

ISBN: (纸本)9798350305418

With the popularity of languages and the increasing demand for language translation, the optimization of artificial intelligence scoring algorithm for English translation has become the urgent need of the translation market. Based on the deep neural network(DNN) algorithm, it continuously integrates modern technology to realize information collection, storage, analysis and application, which promotes the deep integration of the DNN algorithm in the field of intelligent identification and expands the application scope of artificial intelligence identification technology. This paper focuses on how to efficiently optimize the model of English translation artificial intelligence scoring algorithm by DNN, specifically studying the DNN model, by studying the mathematical model and algorithm flow of the DNN optimization algorithm, and the basic principle of the DNN model;Based on the DNN model, a learning ability analysis model of English translation algorithm is proposed. Experiments show that this model can optimize the artificial intelligence scoring algorithm of English translation. © 2023 IEEE.

关键词： Artificial intelligence scoring algorithm DNN English translation

来源：评论

学校读者我要写书评

暂无评论

IoMT-Enabled Smart Healthcare: State-of-the-Art, Security and Future Directions 14

IoMT-Enabled Smart Healthcare: State-of-the-Art, Security an...

引用

14th International Conference on Information and Knowledge Technology, IKT 2023

作者： Tripathi, Shivam Makwana, Vatsalkumar Kumhar, Malaram Trivedi, Harshal Bhatia, Jitendra Tanwar, Sudeep Shahinzadeh, Hossein L. J. Institute of Engineering and Technology Department of Information Technology Ahmedabad India L. J. Institute of Engineering and Technology Department of Computer Engineering Ahmedabad India Nirma University Department of Computer Science Engineering Institute of Technology Ahmedabad India CEO Tushker AI Ahmedabad India Amirkabir University of Technology Department of Electrical Engineering Tehran Iran

ISBN: (纸本)9798350349412

The Internet of Medical Things (IoMT) has become a game-changing force in the healthcare sector, revolutionizing the management of patient care and the provision of medical services. IoMT improves patient monitoring, diagnoses, therapy, and overall healthcare management by using the potential of linked medical devices, wearable sensors, and healthcare systems. This paper thoroughly analyzes recent research trends in computing architectures and communication technologies in IoMT-enabled smart healthcare. We examine the most recent advancements in IoMT technology and applications, emphasizing how they affect patient outcomes, healthcare expenditures, and quality of care. We explore some of the most important areas of IoMT research, such as remote patient monitoring, real-time data analytics, and integration with cutting-edge technology such as machine learning (ml) and deep learning (DL). Furthermore, we explore the use cases in IoT-based healthcare, data analytics techniques, and challenges in IoMT. Security and privacy measures in IoMT are also discussed in the paper. The paper concluded with open discussion and future research directions in IoT-based healthcare. © 2023 IEEE.

关键词： Fog computing

来源：评论

学校读者我要写书评

暂无评论

Clear Text Visibility: A CNN-LSTM Approach for Frame of Interest Identification in Book Flipping Videos

Clear Text Visibility: A CNN-LSTM Approach for Frame of Inte...

引用

Modeling, Simulation & Intelligent Computing (MoSICom), International Conference on

作者： Gaurav Buddhawar Dharin Dave Krupa Jariwala Chiranjoy Chattopadhyay Dept. of Computer Science and Engineering Sardar Vallabhbhai National Institute of Technology Surat India Dept. of Computer Science and Engineering (AI & ML) Pimpri Chinchwad College of Engineering Pune India School of Computing and Data Sciences FLAME University Pune India

ISBN: (数字)9798331533311

ISBN: (纸本)9798331533328

Book flipping videos present a distinctive challenge for information extraction, requiring the identification of frames with clear text visibility during dynamic page turns. This paper introduces a novel approach to frame classification in book flipping videos, leveraging the combined effect of Convolutional Neural Networks (CNNs) and Long Short-Term Memory networks (LSTMs). In particular, a CNN-LSTM model is employed, where a pre-trained CNN captures spatial features, and LSTMs extend the model’s capability to distinguish temporal dependencies critical for page turn detection. Unlike conventional methods, this approach processes entire video sequences, enabling the model to learn intricate Frame of Interest (FoI) detection features indicative of page turns and text visibility. The proposed model offers improved interpretability, effectively classifying frames without compromising transparency. The research opens avenues for automated book digitization, digital library creation, and educational technology applications, enhancing information extraction and reliability from book flipping videos.

关键词： Digital transformation Video sequences Educational technology Information retrieval Feature extraction Libraries Convolutional neural networks Reliability Long short term memory Videos

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：