检索结果-内蒙古大学图书馆

IEEE Games Media Entertainment (GEM)

作者： Hasindu Ramanayake Iresh Bandara School of Computing Informatics Institute of Technology Colombo Sri Lanka The Center for Ubiquitous Computing (UBICOMP) University of Oulu Oulu Finland

ISBN: (数字)9798350374537

ISBN: (纸本)9798350374544

This study aims to introduce a novel methodology for synthesizing emotional expressions in virtual characters, designed to adapt dynamic environmental transitions in real time, thereby enhancing interactive realism and responsiveness in locomotion systems. Emotions, which serve as a universal form of communication, profoundly impact human behaviour to drive instantaneous responses for all time scales of reality. Despite recent studies on motion synthesis, there is still an elusive connection between virtual characters and environments to explore as much as the fidelity needs to grow in both digital and virtual worlds. As a matter of fact, we put forth an innovative approach that leverages psychological frameworks to establish correlations between the character and its environment. This approach is reinforced by a novel deep-learning architecture designed to execute multiple modalities and tasks within a singular network concurrently. Replicating real-world properties in our work will lead to a fascinating initiative that attempts to give these virtual creations the ability to mimic, express, and evoke emotions in virtual environments. An observational study using standard motion synthesis criteria confirmed the system's effectiveness, revealing substantial improvements in performance and accuracy. Eventually, future approaches have been advocated in the context of multi-agent interactions to push the prevailing realism into the next generation. The study has determined that research aims were fulfilled within a challenging process, producing promising results for future applications in this research domain.

关键词： Scalability Virtual environments Psychology Games Production Aerodynamics Real-time systems

来源：评论

学校读者我要写书评

暂无评论

Improving stability of detected topics in learning analytics 45

Improving stability of detected topics in learning analytics

引用

45th Jubilee International Convention on Information, Communication and Electronic Technology, MIPRO 2022

作者： Rako, S. Simic, D. Slibar, B. Gusic-Mundar, J. University of Zagreb University Computing Center Zagreb Croatia University of Zagreb Faculty of Organization and Informatics Varaždin Croatia

ISBN: (纸本)9789532331035

Due to exponential increase in the number of published research papers, it is becoming more and more difficult to review current research in any discipline. It is especially true for disciplines undergoing fast development like learning analytics and educational data mining. Natural language processing offers opportunity to facilitate extraction of information/knowledge from a large volume of documents. In an attempt to organize papers on application of learning analytics and educational data mining for supporting self-regulated learning, we have used Latent Dirichlet Allocation to identify topics in a set of 107 research papers. Results of Latent Dirichlet Allocation depend on the initial choice of a random number generator seed, and replicated runs do not result in the same topics. Stability of allocation of documents to topics was analyzed using Jaccard similarity coefficient. The proposed method enables identifying stabile clusters of research papers sharing the same topics. © 2022 Croatian Society MIPRO.

关键词： Data mining

来源：评论

学校读者我要写书评

暂无评论

Dialogue Management as Graph Transformations 12th

Dialogue Management as Graph Transformations

引用

12th International Workshop on Spoken Dialogue System Technology, IWSDS 2021

作者： Walker, Nicholas Thomas Dahl, Torbjørn Lison, Pierre Norwegian Computing Center Postboks 114 Blindern Oslo0314 Norway Department of Informatics University of Oslo Oslo Norway

ISBN: (纸本)9789811955372

We present ongoing work on a new dialogue management framework using graphs as core representation for the current dialogue state. Dialogue management tasks such as state tracking and action selection are framed as sequences of graph transformations that repeatedly update this graph based on incoming observations. Those graph transformations are expressed using a graph query language, making it possible to specify all dialogue management operations through a unified, declarative syntax. We argue that graphs are particularly well suited to model the dialogue state of complex, open-ended domains. In contrast to traditional dialogue state representations that are limited to fixed, predefined slots, graphs can naturally express dialogue domains with rich relational structures and variable numbers of entities to track. We describe how dialogue state tracking and action selection can be modelled in such graph-centric view of dialogue management, using either handcrafted rules or data-driven models. We also briefly discuss how to account for some aspects of dialogue management such as uncertainties, incremental inputs and contextual knowledge. Finally, we describe a proof-of-concept study of this dialogue management framework in a human–robot interaction scenario. © 2022, The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd.

关键词： Graph Databases

来源：评论

学校读者我要写书评

暂无评论

Can we trust the evaluation on ChatGPT? 3

Can we trust the evaluation on ChatGPT?

引用

3rd Workshop on Trustworthy Natural Language Processing, TrustNLP 2023, co-located with ACL 2023

作者： Aiyappa, Rachith An, Jisun Kwak, Haewoon Ahn, Yong-Yeol Center for Complex Networks & Systems Luddy School of Informatics Computing & Engineering United States Indiana University Network Science Institute Indiana University BloomingtonIN47408 United States

ISBN: (纸本)9781959429869

ChatGPT, the first large language model with mass adoption, has demonstrated remarkable performance in numerous natural language tasks. Despite its evident usefulness, evaluating ChatGPT's performance in diverse problem domains remains challenging due to the closed nature of the model and its continuous updates via Reinforcement Learning from Human Feedback (RLHF). We highlight the issue of data contamination in ChatGPT evaluations, with a case study in stance detection. We discuss the challenge of preventing data contamination and ensuring fair model evaluation in the age of closed and continuously trained models. © 2023 Proceedings of the Annual Meeting of the Association for Computational Linguistics. All rights reserved.

关键词： Reinforcement learning

来源：评论

学校读者我要写书评

暂无评论

Breast Cancer Prediction Using Classical Machine Learning Models 6

Breast Cancer Prediction Using Classical Machine Learning Mo...

引用

6th International Conference on Intelligent Robotics and Control Engineering, IRCE 2023

作者： Almazrouei, Khawla Nassif, Ali Bou Autonomous Robotics Research Center UAE Technology Innovation Institute Abu Dhabi United Arab Emirates College of Computing and Informatics University of Sharjah Sharjah United Arab Emirates

ISBN: (纸本)9798350326338

Machine Learning is an emerging topic of study nowadays. Due to their extensive capacity for data exploration and the discovery of illuminating patterns, machine learning, and data mining techniques have become extremely popular. Data mining techniques are also used in the healthcare industry and have proven particularly effective for cancer prognosis and prediction. Breast cancer is the second most significant factor in women's cancer-related death rates. In that situation, early detection of benign and malignant stages of breast cancer can enable the patient, and the doctor administers the appropriate treatments. In this study, a breast cancer dataset of 569 instances and 32 features were chosen to predict the stages of breast cancer. Four classification algorithms are employed: Support Vector Machine, K Nearest Neighbor, Multi layer Perceptron, and Radial Basis Neural Network. The algorithm's performance regarding Accuracy, Precision, Recall, and Roc Area are compared, and it is found that SVM outperformed the others with an accuracy rate of 98.05%. The paper also demonstrated how feature selection strategies might shorten training times before proposing a reliable and universal model for predicting the breast cancer stage. © 2023 IEEE.

关键词： Multilayer neural networks

来源：评论

学校读者我要写书评

暂无评论

A Supervised Model to Detect Suspicious Activities in the Bitcoin Network 2nd

A Supervised Model to Detect Suspicious Activities in the Bi...

引用

2nd International Conference on Emerging Technologies and Intelligent Systems, ICETIS 2022

作者： Al-Hashedi, Khaled Gubran Magalingam, Pritheega Maarop, Nurazean Samy, Ganthan Narayana Rahim, Fiza Binti Abdul Shanmugam, Mohana Hasan, Mohammad Kamrul Advanced Informatics Department Razak Faculty of Technology and Informatics Universiti Teknologi Malaysia Kuala Lumpur Malaysia College of Computing and Informatics Universiti Tenaga Nasional Malaysia Selangor Kajang43000 Malaysia Center for Cyber Security Faculty of Information Science and Technology Universiti Kebangsaan Malaysia Selangor Bangi43600 Malaysia

ISBN: (纸本)9783031252730

Shortly after its official launch in 2009, Bitcoin has gained rapid popularity worldwide, which in return attracted a variety of people especially malicious attackers, who get the advantage of its pseudo-anonymity to institute un-traceable threats, scams, and criminal activities. Recently, some Bitcoin thefts have been reported costing millions of dollars, causing serious harm and losses to innocent users or companies that lead them to declare bankruptcy. One of the main characteristics of Bitcoin is its anonymity, which makes Bitcoin the preferred choice for criminals to perform illicit activities that pose difficulties for law enforcement and financial authorities to identify suspicious behavior, making the existing fraud detection systems ineffective. In this paper, we propose a model for detecting suspicious activities in the Bitcoin network. We first construct a labeled dataset by collecting a set of illicit transactions from public online Bitcoin forums, as well as datasets from prior research. Next, a verification and filtration process has been performed to verify the gathered illicit transactions with the original dataset and manually marked them as either legal or illegal. Additionally, a new set of features that are based on time-slice was extracted, the skewed dataset was balanced, and three supervised classifiers (LR, NB, and ANN) were used for evaluating the proposed model. Finally, our findings found that the ANN classifier achieved the best performer among others, which attained Precision, Recall, F1 scores, and AUC of 95.2%, 88.7%, 89.8%, and 91.2% respectively. The performance of the supervised classifiers has significantly improved after balancing the training set. © 2023, The Author(s), under exclusive license to Springer Nature Switzerland AG.

关键词： Machine learning

来源：评论

学校读者我要写书评

暂无评论

Holoported Characters: Real-Time Free-Viewpoint Rendering of Humans from Sparse RGB Cameras

Holoported Characters: Real-Time Free-Viewpoint Rendering of...

引用

Conference on Computer Vision and Pattern Recognition (CVPR)

作者： Ashwath Shetty Marc Habermann Guoxing Sun Diogo Luvizon Vladislav Golyanik Christian Theobalt Saarland Informatics Campus Max Planck Institute for Informatics Saarland University Saarbrücken Research Center for Visual Computing Interaction and AI

ISBN: (数字)9798350353006

ISBN: (纸本)9798350353013

We present the first approach to render highly realistic free-viewpoint videos of a human actor in general apparel, from sparse multi-view recording to display, in real-time at an unprecedented 4K resolution. At inference, our method only requires four camera views of the moving actor and the respective 3D skeletal pose. It handles actors in wide clothing, and reproduces even fine-scale dynamic detail, e.g. clothing wrinkles, face expressions, and hand gestures. At training time, our learning-based approach expects dense multi- view video and a rigged static surface scan of the actor. Our method comprises three main stages. Stage 1 is a skeleton-driven neural approach for high-quality capture of the detailed dynamic mesh geometry. Stage 2 is a novel solution to create a view-dependent texture using four test-time camera views as input. Finally, stage 3 comprises a new image-based refinement network rendering the final 4K image given the output from the previous stages. Our approach establishes a new benchmark for real-time rendering resolution and quality using sparse input camera views, unlocking possibilities for immersive telepresence. Code and data is available on our project page.

关键词： Training Three-dimensional displays Telepresence Runtime Clothing Cameras Rendering (computer graphics)

来源：评论

学校读者我要写书评

暂无评论

Enhancing Diverging-Wave Ultrasound Imaging with the Iterative Adaptive Approach

Enhancing Diverging-Wave Ultrasound Imaging with the Iterati...

引用

IEEE International Symposium on Applications of Ferroelectrics (ISAF)

作者： Mahsa Sotoodeh Ziksari Sven Peter Näsholm Andreas Austeng Are Charles Jensen Department of Informatics University of Oslo Oslo Norway Norwegian Computing Center (NR) Oslo Norway

ISBN: (数字)9798350371901

ISBN: (纸本)9798350371918

Diverging-wave transmission is a technique used to enhance temporal resolution, but this comes with inherent limitations in terms of image quality. To address these, various specialized and adaptive beamforming methods have been proposed. In the current study, we demonstrate the feasibility of applying the Iterative Adaptive beamforming Approach (IAA) to coherently compounded diverging-wave transmission data – hence mimicking the previously demonstrated focused-transmit setting. This is done on both synthetic simulated and recorded phantom data. The results demonstrate that the proposed method offers comparable contrast and significantly better resolution than the delay-and-sum (DAS) beamformer applied to the same dataset. Moreover, we demonstrate that the IAA method is also applicable to single-transmit diverging waves. Although this setting is quite different from the original IAA regime, the method produces images of similar or better quality than DAS.

关键词： Image quality Image resolution Ultrasonic imaging Array signal processing MIMICs Imaging Phantoms Iterative methods Image reconstruction Frequency control

来源：评论

学校读者我要写书评

暂无评论

GIGA: Generalizable Sparse Image-driven Gaussian Avatars

arXiv

引用

arXiv 2025年

作者： Zubekhin, Anton Zhu, Heming Gotardo, Paulo Beeler, Thabo Habermann, Marc Theobalt, Christian Max Planck Institute for Informatics Saarland Informatics Campus Germany Saarbrücken Research Center for Visual Computing Interaction and AI Germany Google Switzerland

Driving a high-quality and photorealistic full-body human avatar, from only a few RGB cameras, is a challenging problem that has become increasingly relevant with emerging virtual reality technologies. To democratize such technology, a promising solution may be a generalizable method that takes sparse multi-view images of an unseen person and then generates photoreal free-view renderings of such identity. However, the current state of the art is not scalable to very large datasets and, thus, lacks in diversity and photorealism. To address this problem, we propose a novel, generalizable full-body model for rendering photoreal humans in free viewpoint, as driven by sparse multi-view video. For the first time in literature, our model can scale up training to thousands of subjects while maintaining high photorealism. At the core, we introduce a MultiHeadUNet architecture, which takes sparse multi-view images in texture space as input and predicts Gaussian primitives represented as 2D texels on top of a human body mesh. Importantly, we represent sparse-view image information, body shape, and the Gaussian parameters in 2D so that we can design a deep and scalable architecture entirely based on 2D convolutions and attention mechanisms. At test time, our method synthesizes an articulated 3D Gaussian-based avatar from as few as four input views and a tracked body template for unseen identities. Our method excels over prior works by a significant margin in terms of cross-subject generalization capability as well as photorealism. The project page is available at ***/projects/GIGA. © 2025, CC BY-NC-ND.

关键词： Gaussian distribution

来源：评论

学校读者我要写书评

暂无评论

Private Continuous Survival Analysis with Distributed Multi-Site Data

Private Continuous Survival Analysis with Distributed Multi-...

引用

2023 IEEE International Conference on Big Data, BigData 2023

作者： Bonomi, Luca Lionts, Marilyn Fan, Liyue Vanderbilt University Medical Center Dept. Biomedical Informatics NashvilleTN United States Vanderbilt University Dept. Computer Science NashvilleTN United States University of North Carolina College of Computing and Informatics CharlotteNC United States

ISBN: (纸本)9798350324457

Effective disease surveillance systems require large-scale epidemiological data to improve health outcomes and quality of care for the general population. As data may be limited within a single site, multi-site data (e.g., from a number of local/regional health systems) need to be considered. Leveraging distributed data across multiple sites for epidemiological analysis poses significant challenges. Due to the sensitive nature of epidemiological data, it is imperative to design distributed solutions that provide strong privacy protections. Current privacy solutions often assume a central site, which is responsible for aggregating the distributed data and applying privacy protection before sharing the results (e.g., aggregation via secure primitives and differential privacy for sharing aggregate results). However, identifying such a central site may be difficult in practice and relying on a central site may introduce potential vulnerabilities (e.g., single point of failure). Furthermore, to support clinical interventions and inform policy decisions in a timely manner, epidemiological analysis need to reflect dynamic changes in the data. Yet, existing distributed privacy-protecting approaches were largely designed for static data (e.g., one-time data sharing) and cannot fulfill dynamic data requirements. In this work, we propose a privacy-protecting approach that supports the sharing of dynamic epidemiological analysis and provides strong privacy protection in a decentralized manner. We apply our solution in continuous survival analysis using the Kaplan-Meier estimation model while providing differential privacy protection. Our evaluations on a real dataset containing COVID-19 cases show that our method provides highly usable results. © 2023 IEEE.

关键词： Data Privacy Distributed Data Survival Analysis

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：