检索结果-内蒙古大学图书馆

IEEE International Conference on Image Processing

作者： Jian Lin Xueting Liu Chengze Li Minshan Xie Tien-Tsin Wong School of Computing and Information Sciences Saint Francis University Department of Computer Science and Engineering The Chinese University of Hong Kong Department of Data Science and Artificial Intelligence Monash University

ISBN: (数字)9798350349399

ISBN: (纸本)9798350349405

While manga is a popular entertainment form, creating manga is tedious, especially adding screentones to the created sketch, namely manga screening. Unfortunately, there is no existing method that tailors for automatic manga screening, probably due to the difficulty in generating shaded high-frequency screentones of high-quality. Classic manga screening approaches generally require user input to provide screentone exemplars or a reference manga image. Recent deep learning models enable automatic generation by learning from a large-scale dataset. However, the state-of-the-art models still fail to generate high-quality shaded screentones due to the lack of a tailored model and high-quality manga training data. In this paper, we propose a novel sketch-to-manga framework that first generates a color illustration from the sketch and then generates a screentoned manga based on the intensity guidance. Our method significantly outperforms existing methods in generating high-quality manga with shaded high-frequency screentones.

关键词： Deep learning Image color analysis Training data Entertainment industry Diffusion models Animation data models

来源：评论

学校读者我要写书评

暂无评论

The Necessary Requirements for Utilizing AI in data Access Setting 15

The Necessary Requirements for Utilizing AI in Data Access S...

引用

15th International Conference on Computing Communication and Networking Technologies, ICCCNT 2024

作者： Swapnil, K. Sharma, Bhuvnesh Singh, Manpreet Yadav, Rakesh Kumar Devaraj Verma, C. Umarani, K. Vishwakarma Institute of Technology Department of Artificial Intelligence &data Science Pune India Vivekananda Global University Department of Mechanical Engineering Jaipur India Maharishi University of Technology Maharishi School of Engineering and Technology Uttar Pradesh India Faculty of Engineering and Technology Department of Computer Science and Engineering Karnataka Ramnagar District562112 India Chitkara University Institute of Engineering and technology Chitkara University Centre of Interdisciplinary Research in Business and Technology Punjab India Prince Shri Venkateshwara Padmavathy Engineering College Department of Computer Science and Engineering Chennai India

ISBN: (纸本)9798350370249

AI has been an increasing number of adopted in records get admission to settings to improve the overall performance of statistics get admission to and extract significant insights, using extended productiveness and value-effectiveness. However, the success implementation of AI calls for positive essential requirements to be met, together with facts accuracy, connectivity, security, and scalability. First, facts accuracy is critical for AI applications in facts get entry to settings. Information gathered need to be of excessive quality and accuracy, as AI working on erroneous facts can yield unreliable and probably harmful effects. Records should be of suitable granularity, taking pictures the most relevant elements, to make certain a complete and steady training performance.2nd, linked statistics assets, networks, and services are had to leverage the capability of AI in information get entry to settings. This gives an interconnected and up to date dataset at scale, while requiring included answers for uninterrupted get entry to. Moreover, single sign-on and alertness orchestration abilities are vital to decorate user revel in. © 2024 IEEE.

关键词： Connectivity Gathered Increasing Productiveness Security

来源：评论

学校读者我要写书评

暂无评论

Pheromone Matrix Eigenvalue-based Convergence Assessment Indicator for Ant Colony Optimization 11

Pheromone Matrix Eigenvalue-based Convergence Assessment Ind...

引用

11th International Conference on Machine Intelligence Theory and Applications, MiTA 2024

作者： Li, Xu Shi, Lin Li, Jian-Yu Xu, Jing Zhan, Zhi-Hui South China University of Technology School of Computer Science and Engineering Guangzhou510006 China Guangdong University of Finance and Economics Research Center of Intelligent Computing & Big Data Technology Guangdong Guangzhou510320 China Nankai University College of Artificial Intelligence Tianjin300350 China

ISBN: (纸本)9798350385670

As a crucial member of the evolutionary computation (EC), ant colony optimization (ACO) converges to the optimal result through a continuous process of iterative optimization. However, determining the convergence status of an ant colony remains a challenging task without well-established methods. To this end, this paper proposes a global evaluation indicator - average absolute eigenvalue of pheromone matrix (AAEPM) by constructing a pheromone matrix and analyzing its eigenvalue variations. This indicator aims to assess the convergence status of the ant colony during the evolutionary process. Experimental results show that the AAEPM is correlated with the convergence status of ACO and is not significantly influenced by algorithm parameters and problem scale. © 2024 IEEE.

关键词： Anthropogenic

来源：评论

学校读者我要写书评

暂无评论

Comparative Analysis of Predictive Models for Post-Surgery Kyphosis Persistence: Using Machine Learning Techniques for Clinical Prognosis

Comparative Analysis of Predictive Models for Post-Surgery K...

引用

Advancement in Computation & computer Technologies (InCACCT), International Conference on

作者： Harshvardhan Tiwari Neel Pandey Baswaraju Swathi School of Tech New Zealand Skills and Education Group Auckland New Zealand Department of Computer Science and Engineering (Data Science) New Horizon College of Engineering Bangalore India

ISBN: (数字)9798350371314

ISBN: (纸本)9798350371321

Kyphosis, an abnormal curvature of the upper spine, frequently necessitates surgical treatment, particularly in adolescence. Yet, assessing the duration of the illness after treatment is difficult however essential for predicting the patient’s prognosis. The study evaluated the performance of various machine learning models on the Kyphosis classification task using K-fold cross-validation using 5 and 10 folds. The dataset was divided into 5 subsets with 5 folds and 10 subsets with 10 folds, allowing for accurate predictions on unseen data. K-Nearest Neighbours was the best-performing model, with an average cross-validation score of around 0.864. Random Forest and Support Vector Machine followed, achieving impressive scores of around 0.854 and 0.842, respectively. Naive Bayes showed slightly lesser performance, while Logistic Regression showed the lowest performance. In the 10 splits scenario, KNearest Neighbours continued to perform well, with an average cross-validation score of around 0.887. The results offer useful insights into how different models perform under various cross-validation methods, helping in the selection of suitable algorithms for predictive modelling. The study highlights KNearest Neighbours as the most reliable performer in both 5 splits and 10 splits. Basic ML algorithms-based models are competitive with deep learning models with accuracy rates ranging from 74% to 88%, offering a balance between performance and computational complexity. Further the integration of these models into clinical practice to provide dependable predictions regarding the persistence of Kyphosis after surgery, particularly in settings where computational resources are limited was suggested. This offers vital information for healthcare professionals to predict postoperative results with precision. Our research highlights the importance of utilizing machine learning as a key tool in tackling a variety of biological challenges.

关键词： Support vector machines Computational modeling Surgery Medical services Predictive models Prediction algorithms Reliability

来源：评论

学校读者我要写书评

暂无评论

Selective Temporal Knowledge Graph Reasoning

arXiv

引用

arXiv 2024年

作者： Hou, Zhongni Jin, Xiaolong Li, Zixuan Bai, Long Guo, Jiafeng Cheng, Xueqi School of Computer Science and Technology University of Chinese Academy of Sciences China Key Laboratory of Network Data Science and Technology Institute of Computing Technology Chinese Academy of Sciences China

Temporal Knowledge Graph (TKG), which characterizes temporally evolving facts in the form of (subject, relation, object, timestamp), has attracted much attention recently. TKG reasoning aims to predict future facts based on given historical ones. However, existing TKG reasoning models are unable to abstain from predictions they are uncertain, which will inevitably bring risks in real-world applications. Thus, in this paper, we propose an abstention mechanism for TKG reasoning, which helps the existing models make selective, instead of indiscriminate, predictions. Specifically, we develop a confidence estimator, called Confidence Estimator with History (CEHis), to enable the existing TKG reasoning models to first estimate their confidence in making predictions, and then abstain from those with low confidence. To do so, CEHis takes two kinds of information into consideration, namely, the certainty of the current prediction and the accuracy of historical predictions. Experiments with representative TKG reasoning models on two benchmark datasets demonstrate the effectiveness of the proposed CEHis. Copyright © 2024, The Authors. All rights reserved.

关键词： Knowledge graph

来源：评论

学校读者我要写书评

暂无评论

Language-agnostic Zero-shot Entity Linking via Entity Prototype Network

Language-agnostic Zero-shot Entity Linking via Entity Protot...

引用

IEEE International Conference on Big Knowledge (ICBK)

作者： Haihong Yang Ming Qin Qiang Zhang Renjun Xu College of Computer Science and Technology Zhejiang University Hangzhou China School of Software Technology Zhejiang University Hangzhou China Center for Data Science Zhejiang University Hangzhou China

ISBN: (数字)9798331508821

ISBN: (纸本)9798331508838

Following the recent trend of language understanding, we introduce a novel approach for constructing a general-purpose entity linking system by learning entity prototypes. Our model, the Entity Prototype Network (EPN), generates language-agnostic entity prototypes by reading entity descriptions and aligning diverse surface forms in different languages to the corresponding entity prototype in a unified vector space. To validate our core assumption of learnable language-agnostic entity prototypes, we define a new Zero-shot Cross-lingual Entity Linking (ZXEL) task. Additionally, we propose a simple yet effective auxiliary task termed Entity Identification, which enhances our model’s ability to capture both intra-entity similarity and inter-entity dissimilarity. We evaluate our model on the new task in both classic and generalized settings. Experimental results demonstrate consistent improvements (exceeding 30% on average) over competitive baselines. An ablation study confirms the necessity of our model design and highlights the impact of the proposed auxiliary task.

关键词： Translation Prototypes Knowledge graphs Market research Vectors

来源：评论

学校读者我要写书评

暂无评论

Navigating Conflicting Views: Harnessing Trust for Learning

arXiv

引用

arXiv 2024年

作者： Lu, Jueqing Buntine, Wray Qi, Yuanyuan Dipnall, Joanna Gabbe, Belinda Du, Lan Department of Data Science & AI Monash University Australia College of Engineering and Computer Science VinUniversity Viet Nam School of Public Health and Preventive Medicine Monash University Australia

Resolving conflicts is critical for improving the reliability of multi-view classification. While prior work focuses on learning consistent and informative representations across views, it often assumes perfect alignment and equal importance of all views, an assumption rarely met in real-world scenarios, as some views may express distinct information. To address this, we develop a computational trust-based discounting method that enhances the Evidential Multi-view framework by accounting for the instance-wise reliability of each view through a probability-sensitive trust mechanism. We evaluate our method on six real-world datasets using Top-1 Accuracy, Fleiss’ Kappa, and a new metric, Multi-View Agreement with Ground Truth, to assess prediction reliability. We also assess the effectiveness of uncertainty in indicating prediction correctness via AUROC. Additionally, we test the scalability of our method through end-to-end training on a large-scale dataset. The experimental results show that computational trust can effectively resolve conflicts, paving the way for more reliable multi-view classification models in real-world applications. Codes available at: https://***/OverfitFlow/Trust4Conflict © 2024, CC BY.

关键词： Uncertainty analysis

来源：评论

学校读者我要写书评

暂无评论

Deep Learning-Based Buoyancy Prediction for Intelligent Mussel Farm Monitoring

Deep Learning-Based Buoyancy Prediction for Intelligent Muss...

引用

International Conference on Image and Vision Computing New Zealand, IVCNZ

作者： Carl McMillan Junhong Zhao Bing Xue Ross Vennell Mengjie Zhang Centre for Data Science and Artificial Intelligence School of Engineering and Computer Science Victoria University of Wellington New Zealand Coastal and Freshwater Group Cawthron Institute Nelson New Zealand

ISBN: (数字)9798331518776

ISBN: (纸本)9798331518783

Greenshell mussel farming is the largest contributor to export earnings in Aotearoa New Zealand's (NZ) aquaculture industry. Floats are essential to mussel farm structures, keeping tons of mussels suspended in the water. Monitoring float buoyancy is essential for minimising the loss of mussels due to sinking or excessive energy transfer to the mussels. We propose a novel transformer-based neural network architecture for float buoyancy prediction in NZ's greenshell mussel farms. The model extends a vision transformer to incorporate relationships between adjacent floats, enhancing prediction performance. Results show improved accuracy over baseline methods, suggesting potential for integration into a cost-effective, computer vision-based monitoring system for mussel farm buoyancy at scale. This approach could address the challenges of monitoring buoyancy in expanding coastal and open ocean aquaculture operations.

关键词： Accuracy Computational modeling Oceans Neural networks Sea measurements Buoyancy Predictive models Transformers Monitoring Aquaculture

来源：评论

学校读者我要写书评

暂无评论

A Reinforcement Learning Agent Controlled Multi-branch Small Object Detection Framework

A Reinforcement Learning Agent Controlled Multi-branch Small...

引用

International Conference on Acoustics, Speech, and Signal Processing (ICASSP)

作者： Junkun Hong Yitian Long Yueyi Luo Liujie Hua Jun Long Qianqian Qi School of Computer Science and Engineering Central South University Changsha China Data Science Institute Vanderbilt University Nashville Tennessee United States School of Mathematics and Statistics Central South University Changsha China Big Data Institution Central South University Changsha China

ISBN: (数字)9798350368741

ISBN: (纸本)9798350368758

The past few years have witnessed the immense development of small object detection, which is aimed at detecting size-limited targets in high-resolution images. The prevailing methods focus on extracting fine-grained information by expanding the receptive fields and then generating the potential small object region. However, these solutions inevitably add sophisticated detectors and extra learning components, which incur time-consuming and computation-costing. Meanwhile, we observe that it’s suboptimal to extract fine features in such a generic way. To alleviate the issues, we propose a multi-branch small object detection framework with a regular-scale detection branch and a small-scale detection branch. Specifically, we design and pre-train a reinforcement learning agent to control feature extractors in both branches according to the results of small object areas. Moreover, we present a region clipping algorithm to rebuild the small object to regular size, which can be input into a mature detector directly. The extensive experiments on COCO, VisDrone, SODA-D, and our collecting TVDS datasets demonstrate our method outperforms the state-of-the-art methods in several metrics.

关键词： Measurement Pipelines Signal processing algorithms Object detection Reinforcement learning Detectors Signal processing Feature extraction data mining Speech processing

来源：评论

学校读者我要写书评

暂无评论

Improving Medical Imaging Diagnostics with Deep Convolutional Networks for Early Detection and Treatment

Improving Medical Imaging Diagnostics with Deep Convolutiona...

引用

2025 International Conference on Computing for Sustainability and Intelligent Future, COMP-SIF 2025

作者： Krishnakumar, K. Gurrapu, Omprakash Reshma, P. Krishnammal, P. Muthu Khan, Huma Qamar Amane, Ninad Saveetha School of Engineering Chennai India Volvo Trucks North America NC United States Dhanalakshmi Srinivasan College of Engineering Department of Artificial Intelligence and Data Science Coimbatore India VIT-AP University Amaravati India Joginpally B R Engineering College Department of Computer Science and Engineering Hyderabad India Bms Institute of Technology and Management Department of Ai&ml Bangalore India

ISBN: (纸本)9798331538538

Diagnostic imaging is an important technique which helps in diagnosing several diseases at the initial stage in order to take necessary treatments. Nevertheless, conventional techniques tend to have issues with regard to precision, speed, and quantitatively assessing multidimensional medical information. In this paper, we reveal the potential of deep convolutional networks (DCNs) in the field of medical image diagnostic. It means that, with the help of deep learning algorithms DCNs can learn meaningful features of medical images and improve diagnostic accuracy and speed. The near proposed technique involves the use of deep convolutional network model to analyse medical images for diagnosis of diseases such as cancer, heart diseases and neurological disorders. The obtained experimental results demonstrate that the presented model has the highest accuracy of 96.5% with very high sensitivity and specificity coefficients compared to other methods. Further, the interpretability of the proposed DCN model and different aspects of it are discussed to support clinical validation. Through this research, the effectiveness of deep learning in revolutionizing the diagnosis for medical imaging diagnostics while early disease detection and treatment is established. © 2025 IEEE.

关键词： Neuroimaging

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：