检索结果-内蒙古大学图书馆

18th international conference on Interfaces and Human computer Interaction 2024, IHCI 2024, the 17th international conference on Game and Entertainment Technologies 2024, GET 2024 and the 18th international conference on computer graphics, Visualization, computer vision and Image Processing 2024, CGVCVIP 2024, Part of the 18th Multi conference on computer Science and Information Systems 2024, MCCSIS 2024

作者： Mao, Guojun Zhang, Xinchao Fujian Provincial Key Lab of Big Data Mining and Applications Fujian University of Technology Fuzhou350118 China School of Computer Science and Mathematics Fujian University of Technology Fuzhou350118 China

ISBN: (纸本)9789898704580

Camouflaged Object Detection (COD) aims to identify target objects with high similarity to the surrounding environment in complex scenes, and has high application value in military, medical and other fields. this paper proposes a Receptive enhanced based Multi-scale feature fusion network, termed as RMNet. Firstly, it uses the popular pyramid visual network PVTv2 for feature extraction. By overlaying the target area amplification module OAA and deep interaction guidance module DIDM on it, the position ability for camouflaged objects in complex scenes is improved;Secondly, design an enhanced receptive field module REM to enhance the edge perception ability of camouflaged objects;Finally, the feature fusion module MCFM are designed through cross scale hierarchical connections between adjacent levels, effectively improving the accuracy of camouflage object detection. the presented model is evaluated on three highly challenging benchmark datasets, including CAMO, CHAMELEON and COD10K. the evaluation results show that compared with the current popular methods, our method demonstrates superior performance. © 2024 Interfaces and Human computer Interaction. All Rights Reserved.

关键词： Feature extraction

来源：评论

学校读者我要写书评

暂无评论

Bag of World Anchors for Instant Large-Scale Localization

引用

IEEE TRANSACTIONS ON VISUALIZATION AND computer graphics 2023年第11期29卷 4730-4739页

作者： Reyes-Aviles, Fernando Fleck, Philipp Schmalstieg, Dieter Arth, Clemens VRVis Competence Ctr Vienna Vienna Austria Graz Univ Technol Graz Austria

In this work, we present a novel scene description to perform large-scale localization using only geometric constraints. Our work extends compact world anchors with a search data structure to efficiently perform localization and pose estimation of mobile augmented reality devices across multiple platforms (e.g., HoloLens 2, iPad). the algorithm uses a bag-of-words approach to characterize distinct scenes (e.g., rooms). Since the individual scene representations rely on compact geometric (rather than appearance-based) features, the resulting search structure is very lightweight and fast, lending itself to deployment on mobile devices. We present a set of experiments demonstrating the accuracy, performance and scalability of our novel localization method. In addition, we describe several use cases demonstrating how efficient cross-platform localization facilitates sharing of augmented reality experiences.

关键词： Camera localization Correspondence problem 3D registration Augmented Reality computer vision Cross-platform Collaborative Structural modeling

来源：评论

学校读者我要写书评

暂无评论

COMMON AND UNIQUE USER EXPERIENCE PROBLEMS ON E-COMMERCE WEBSITES 17

COMMON AND UNIQUE USER EXPERIENCE PROBLEMS ON E-COMMERCE WEB...

引用

17th international conference on Interfaces and Human computer Interaction 2023, IHCI 2023, the 17th international conference on computer graphics, Visualization, computer vision and Image Processing 2023, CGVCVIP 2023 and 16th international conference on Game and Entertainment Technologies 2023, GET 2023

作者： Hasan, Layla Faculty of Computing Universiti Teknologi Malaysia Johor Malaysia

ISBN: (纸本)9789898704498

this research evaluated the user experience (UX) on the top three most visited e-commerce websites in Malaysia, and identified common and unique issues which affected user experience while interacting with such sites. Specific user experience criteria were used in the evaluation, which was adapted from earlier research, and consists of 40 metrics. these metrics were used to evaluate the utilitarian and hedonic features on the websites. the evaluation was conducted using three methods: quantitative questionnaires, qualitative open-ended questions, and qualitative semi-structured interviews. Twenty students participated in this research and they identified ten common user experience problems related to the utilitarian features (e.g., illogical organisation of the websites' content and lack of presenting detailed information about the products);and three unique user experience problems related to the utilitarian features (i.e., not obvious navigation and lack of suggesting complementary products from other shops). Furthermore, the participants identified four common user experience problems related to the hedonic features (e.g., not visually attractive websites, not creative design, and a lack of desirable design elements which evoke users' emotions and gratitude);they also identified seven unique user experience problems on only one of the tested websites (i.e., the website was not motivating). © 2023 Proceedings of the international conferences on Interfaces and Human computer Interaction 2023, IHCI 2023;computer graphics, Visualization, computer vision and Image Processing 2023, CGVCVIP 2023;and Game and Entertainment Technologies 2023, GET 2023. All rights reserved.

关键词： Human computer interaction

来源：评论

学校读者我要写书评

暂无评论

Randomrewardmechanisms (Loot Boxes): Behavioral Economics As A Regulatory Solution In Brazilian Law 17

Randomrewardmechanisms (Loot Boxes): Behavioral Economics As...

引用

作者： Fantini, Laiane Maris Caetano PUC Minas. Centro DTIBR MG Belo Horizonte Brazil

ISBN: (纸本)9789898704498

the paper seeks to reflect on ways to create an effective regulation by understanding the aspects of random reward mechanisms in video games, also known as loot boxes. Players acquire digital assets of a game through microtransactions, and pays with in-game currency or real money. Among these digital assets, several games also offer loot boxes, which are systems of "surprise boxes" with a certain price but whose reward is unknown to the player (and is only revealed when the box is purchased and opened). A simple way of selling a digital asset has serious implications for players, mainly due to the similarity with gambling in making the reward relegated to chance. Also, adults and children players are encouraged to exhibit addictive behaviors, spending time and money without any control. Brazilian law does not have enough apparatus to protect consumers from these practices and banning loot boxes from the market does not seem to be the solution, as similar mechanisms could arise even worse. this way of marketing digital assets hides the real transaction costs and makes consumers make decisions based on heuristics, and that's why the Behavioral Economic theory can offer better solutions to protect players. © 2023 Proceedings of the international conferences on Interfaces and Human computer Interaction 2023, IHCI 2023;computer graphics, Visualization, computer vision and Image Processing 2023, CGVCVIP 2023;and Game and Entertainment Technologies 2023, GET 2023. All rights reserved.

关键词： Human computer interaction

来源：评论

学校读者我要写书评

暂无评论

DESIGN GUIDELINES FOR PERSONALITY BASED INTERFACES 18

DESIGN GUIDELINES FOR PERSONALITY BASED INTERFACES

引用

作者： Zekry, Dina A. Nagaty, Khaled McKee, Gerard T. The British University in Egypt Cairo El-Sherouk Egypt

ISBN: (纸本)9789898704580

the research presented in this paper addresses the lack of design guidelines for personalization of interfaces in existing interface design principles, which are general in nature. the research investigates the association between users’ dominant personality traits, measured by the Big Five Personality Traits framework, and interface preferences. the research focused on associating five design attributes to the identified dominant traits. the five design attributes are color preferences, information structure, element shape, interface type, and symmetry preferences. Students studying Informatics and computer Science were invited to participate in the study by taking the Big Five Personality Traits test. Segmentation using K-means did not give distinctive personality clusters. Further analysis using Principal Component Analysis (PCA), however, revealed that openness and conscientiousness accounted for over 50% of the variance in the data. the associations were investigated by designing Learning Management System (LMS) based mobile interfaces with attributes corresponding to these two dominant traits. Using the Chi-squared test, the research identified a significant association between the interfaces designed with the openness and conscientiousness proposed attributes and the users’ interface preferences. these findings suggest employing the design attributes as guidelines to facilitate the personalization process for interface designs tailored to openness and conscientiousness personality traits. © 2024 Interfaces and Human computer Interaction. All Rights Reserved.

关键词： Human computer interaction

来源：评论

学校读者我要写书评

暂无评论

MinMax-CAM: Increasing Precision of Explaining Maps by Contrasting Gradient Signals and Regularizing Kernel Usage 17th

MinMax-CAM: Increasing Precision of Explaining Maps by Contr...

引用

17th international Joint conference on computer vision, Imaging and computer graphics theory and Applications (VISIGRAPP) / 17th international conference on computer vision theory and Applications (VISAPP)

作者： David, Lucas Pedrini, Helio Dias, Zanoni Univ Estadual Campinas Campinas SP Brazil

ISBN: (纸本)9783031457241;9783031457258

While various AI explaining methods has been developed in the last decade, the Class Activation Map (CAM) technique (and derivations thereof) has remains a popular alternative today. Highly versatile, and sufficiently simple, CAM-based techniques have been consistently used to explain the predictions of Convolutional Neural Networks over classification problems. However, many studies conducted to assert the precision of the produced explaining maps have done so on single-label multi-class classification problems, where the task is simplified by focusing and framing the objects of interest in the foreground. In this work, we test and evaluate the efficacy of CAM-based techniques over distinct multi-label sets. We find that techniques that were created with single-label classification in mind (such as Grad-CAM, Grad-CAM++ and Score-CAM) will often produce diffuse visualization maps in multi-label scenarios, overstepping the boundaries of their explaining objects of interest onto objects of different classes. We propose a generalization of the Grad-CAM technique for the multi-label scenario that produces more focused explaining maps by maximizing the activation of a class of interest while minimizing the activation of the remaining classes present in the sample. Our results indicate that MinMax-CAM produces more focused explaining maps over different network architectures and datasets. Finally, we present a regularization strategy that encourages sparse positive weights in the last classifying, while penalizing the association between the classification of a class and the occurrence of correlated patterns, resulting in cleaner activation maps.

关键词： computer vision Multi-label Explainable artificial intelligence

来源：评论

学校读者我要写书评

暂无评论

MESH-POLYPDET: SURFACE MESHES BASED POLYP DETECTION IN VIRTUAL COLONOSCOPY 18

MESH-POLYPDET: SURFACE MESHES BASED POLYP DETECTION IN VIRTU...

引用

作者： Ma, Ming Sun, Chenyang Lei, Na Department of Graduate Computer Science and Engineering Katz School of Science and Health Yeshiva University United States School of Software Technology Dalian University of Technology China International School of Information Science & Engineering Dalian University of Technology China

ISBN: (纸本)9789898704580

Colorectal polyps can evolve into colon cancer over time. Early screening or detection of colon polyps using computer-aided detection (CAD) techniques along with removal of the polyps can lower the risk of colon cancer. As an alternative to optimal colonoscopy (OC), virtual colonoscopy (VC) has the advantage of non-invasiveness in the detection of colon polyps. Recent decades have witnessed rapid advancement in VC based polyp detection and false positive (FP) reduction techniques using machine learning or deep learning methods. However, the majority of existing methods are based on the computation using computed tomography (CT) images of patient data, which are in some cases very computationally expensive, particularly in the scenario of necessity of large dataset for model training. In this paper, we propose a novel automatic polyp detection method termed Mesh-PolypDet using the surface meshes of colorectal data. the initial detection of polyp candidates is performed on 3D colon surface meshes, and the subsequent FP reduction is carried out using the visual transformer and conformal mapping. Experiments are conducted to evaluate the performance of the proposed method in comparison with other existing methods. © 2024 Interfaces and Human computer Interaction. All Rights Reserved.

关键词： Conformal mapping

来源：评论

学校读者我要写书评

暂无评论

BONE POROSITY AND thICKNESS EVALUATION BY ULTRASONIC SIGNALS USING PATTERN RECOGNITION 18

BONE POROSITY AND THICKNESS EVALUATION BY ULTRASONIC SIGNALS...

引用

作者： Mamontovs, Maksims Sisojevs, Aleksandrs Tatarinov, Alexey Riga Technical University 6A Ķīpsalas Str. Riga Latvia Institute of Electronics and Computer Science 14 Dzerbenes Str. Riga Latvia

ISBN: (纸本)9789898704580

the fragility of long bones in osteoporosis is linked to a decrease in cortical layer thickness (Cth) and the development of internal porosity (P) within the cortex. this study is dedicated to investigate whether it is possible and how accurately P and Cth can be determined using a pattern recognition methods, utilizing discrete Fourier transform (DFT) analysis of ultrasonic signals transmitted through a soft tissue layer. As input data, ultrasound scans of phantom of cortical bone with different grades of Cth and P were used. Each of the phantom scans was chosen in turn as a test object, while other specimens composed the training set. For analysis, three different existing approaches were applied firstly and the later one was developed by improving the existing ones. the analysis was carried out in three categories: P and Cth ratio, P only and Cth only. Based on the results of each analysis, a method was obtained that predicts P and Cth better than other applied ones. A comparison of the methods for P and Cth determining and choosing the most effective ones are shown. © 2024 Interfaces and Human computer Interaction. All Rights Reserved.

关键词： Discrete Fourier transforms

来源：评论

学校读者我要写书评

暂无评论

AN ANALYSIS OF INTERACTION BEHAVIORS IN PERSONALIZED LEARNING 18

AN ANALYSIS OF INTERACTION BEHAVIORS IN PERSONALIZED LEARNIN...

引用

作者： Cabı, Emine Türkoğlu, Hacer Baskent University Faculty of Education Department of Computer and Instructional Technology Education Ankara Turkey

ISBN: (纸本)9789898704580

the purpose of this study is to examine the relationship between students' participation behaviors in online learning analytics (LA) and their academic performance and self-regulated learning (SRL) in learning analytics-based feedback systems. In the study, 73 university students participated. Moodle was used as a learning management system (LMS) to share digital content, communicate with students, and make evaluations. In the flipped learning (FL) environment, LA including different types of activities was made available to students for four weeks. During this process, students were supported with feedback within the framework of instructional design based on learning analytics. Interaction data, achievement test, and SRL Scale were used as data collection tools. the data obtained in the study were examined using K-Means Clustering Analysis and Mann-Whitney U-test. Cluster analysis was performed to classify students with similar behavioral patterns into a homogeneous group. As a result of the analysis, two clusters were obtained: those with high engagement and those with low engagement. Students' interaction behaviors were determined and the relationship between different interaction behaviors and students' academic success and self-regulation skills was examined. As a result, students with high interaction/engagement spend more time in activities than students with low engagement. Additionally, the academic success and self-regulation score averages (except for help-seeking) of students with high engagement were found to be higher than those with low engagement. When academic achievement and self-regulation skills were compared according to clusters, no statistically significant difference was observed. © 2024 Interfaces and Human computer Interaction. All Rights Reserved.

关键词： Federated learning

来源：评论

学校读者我要写书评

暂无评论

VISUAL ANALYTICS FOR thE MARRIAGE NETWORK IN thE GORYEO DYNASTY, Korea 17

VISUAL ANALYTICS FOR THE MARRIAGE NETWORK IN THE GORYEO DYNA...

引用

作者： Mun, Seongmin Lee, Hyunju Lee, Sangkuk Department of Humanities Research Institute Ajou University Ajou University 206 World cup-ro Korea Republic of Department of History Ajou University Ajou University 206 World cup-ro Korea Republic of

ISBN: (纸本)9789898704498

Political marriage has been a historically productive and important matter to maintain or extend social power. this kind of relationship is also found in Korean history. For instance, in the history of the Goryeo dynasty (i.e., AD 918~ 1392 in Korea), the royal families of the Goryeo dynasty used marriage as a political strategy to maintain or extend their power. Qualitative studies have found that royal families used marriage as a means of maintaining political power, but there is still a lack of quantitative research supporting the outcomes of these studies. For this reason, we adopt computational approaches to advance our understanding of the role of political marriage in the Goryeo dynasty, in Korea. In addition, to simulate the change of power by their origin (i.e., region), we made a visualization system to explore the marriage network of royal families. We note two major findings of this current study. First, the marriage strategies of royal families are divided into three types: (1) exogamous marriages, (2) endogamous marriages, and (3) the combination of exogamous marriages and endogamous marriages. the royal family of Goryeo expanded their external relationship through exogamous marriages and reproduced their power through intra-family marriages. Second, we found that the power of a region was greater when the king was born through wedlock in that region, as opposed to when many marriages were made with the royal family. As a whole, the success of our computational approaches adds to the quantitative evidence for the historical strategies of political marriage in the Goryeo dynasty, in Korea. © 2023 Proceedings of the international conferences on Interfaces and Human computer Interaction 2023, IHCI 2023;computer graphics, Visualization, computer vision and Image Processing 2023, CGVCVIP 2023;and Game and Entertainment Technologies 2023, GET 2023. All rights reserved.

关键词： computer games

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：