检索结果-内蒙古大学图书馆

5th Symposium on Pattern Recognition and Applications, SPRA 2024

作者： Šikić, Franko Vršnak, Donik Lončarić, Sven University of Zagreb Faculty of Electrical Engineering and Computing Zagreb Croatia

ISBN: (数字)9781510688803

ISBN: (纸本)9781510688797

In this paper, we present a survey of deep learning-based methods for the regression of gaze direction vector from head and eye images. We describe in detail numerous published methods with a focus on the input data, architecture of the model, and loss function used to supervise the model. Additionally, we present a list of datasets that can be used to train and evaluate gaze direction regression methods. Furthermore, we noticed that the results reported in the literature are often not comparable one to another due to differences in the validation or even test subsets used. To address this problem, we re-evaluated several methods on the commonly used in-the-wild Gaze360 dataset using the same validation setup. The experimental results show that the latest methods, although claiming state-of-the-art results, significantly underperform compared with some older methods. Finally, we show that the temporal models outperform the static models under static test conditions. Copyright © 2025 SPIE.

关键词： Support vector regression

来源：评论

学校读者我要写书评

暂无评论

SV-data2vec: Guiding Video Representation Learning with Latent Skeleton Targets

SV-data2vec: Guiding Video Representation Learning with Late...

引用

2025 IEEE/CVF Winter Conference on Applications of Computer Vision, WACV 2025

作者： Doždor, Zorana Hrkać, Tomislav Kalafatić, Zoran University of Zagreb Faculty of Electrical Engineering and Computing Croatia

ISBN: (纸本)9798331510831

Recent advancements in action recognition leverage both skeleton and video modalities to achieve state-of-the-art performance. However, due to the challenges of early fusion, which tends to underutilize the strengths of each modality, existing methods often resort to late fusion, consequently leading to more complex designs. Additionally, self-supervised learning approaches utilizing both modalities remain underexplored. In this paper, we introduce SV-data2vec, a novel self-supervised framework for learning from skeleton and video data. Our approach employs a student-teacher architecture, where the teacher network generates contextualized targets based on skeleton data. The student network per-forms a masked prediction task using both skeleton and visual data. Remarkably, after pretraining with both modalities, our method allows for fine-tuning with RGB data alone, achieving results on par with multimodal approaches by effectively learning video representations through skeleton data guidance. Extensive experiments on benchmark datasets NTU RGB+D 60, NTU RGB+D 120, and Toyota Smarthome confirm that our method outperforms existing RGB based state-of-the-art techniques. The code is available at github. com/zoranadozdor/SVdata2vec. © 2025 IEEE.

关键词： Self-supervised learning

来源：评论

学校读者我要写书评

暂无评论

Analysis of User Experience and Task Performance in a Multi-User Cross-Reality Virtual Object Manipulation Task 17

Analysis of User Experience and Task Performance in a Multi-...

引用

17th International Workshop on IMmersive Mixed and Virtual Environment Systems, MMVE 2025, co-located with ACM Multimedia Systems Conference, MMSys 2025

作者： Brzica, Lea Matanović, Filip Vlahović, Sara Pavlin-Bernardić, Nina Skorin-Kapov, Lea University of Zagreb Faculty of Electrical Engineering and Computing Zagreb Croatia University of Zagreb Faculty of Humanities and Social Sciences Zagreb Croatia

ISBN: (纸本)9798400714689

This paper presents a user study (N=40) exploring the impact of different virtual object manipulation techniques on user experience and task performance in a networked cross-reality application. Participants were organized into pairs, with one situated in an Augmented Reality (AR) environment in a real-world lab and the other in its corresponding Virtual Reality (VR) digital twin. The task required participants to take turns moving chess pieces to predetermined positions on a chessboard. The study was organized into six scenarios, each combining one of three manipulation techniques with one of two chess set sizes. Task performance measures included completion time and error rates, while subjective measures encompassed overall Quality of Experience (QoE), usability, co-presence, quality of interaction, and consistency with expectations. Results demonstrated a clear preference for the Pick, Drag, and Snap (PDaS) manipulation technique, which showed significantly faster completion times and higher user satisfaction. This research contributes to the growing body of knowledge on designing effective interaction techniques for cross-reality systems, revealing differences in results and preferences between AR and VR groups, particularly in task completion time. © 2025 Copyright held by the owner/author(s). Publication rights licensed to ACM.

关键词： Virtual environments

来源：评论

学校读者我要写书评

暂无评论

Vessel Registration Number Detection and Recognition System

Vessel Registration Number Detection and Recognition System

引用

2025 IEEE/CVF Winter Conference on Applications of Computer Vision Workshops, WACVW 2025

作者： Fabijanić, Matej Magdalenić, Maja Obradović, Juraj Kapetanović, Nadir Ferreira, Fausto Mišković, Nikola University of Zagreb Faculty of Electrical Engineering and Computing Croatia

ISBN: (纸本)9798331536626

Accurate identification of ships is vital for global trade and maritime security. More specifically, reliable recognition of the vessel registration number helps in orderly navigation and port use for maritime transportation activities. The purpose of this research was to develop and analyze different methods for detecting vessel registration numbers with a monocular camera. We used existing OCR tools available online and modified them to improve their out-of-the-box performance. Three neural network-based text detection methods were developed and tested. All three methods use different text region detection modules for real scenes, while available OCR software is used in all of them for specific character recognition. The methods are compared using standard optical character recognition metrics. The best method that was developed uses a deep neural network model to detect a vessel, then another fine-tuned neural network model to detect a text region, coupled with deterministic image processing methods to improve picture quality. In conclusion, with an accuracy of 72.69% recognized characters of vessel registrations using a limited image dataset, we achieve satisfactory results. Further effort focusing on collecting and annotating diverse maritime environment datasets featuring footage of vessels with visible registration numbers is needed to have a more reliable and robust system. © 2025 IEEE.

关键词： Optical character recognition

来源：评论

学校读者我要写书评

暂无评论

Towards a Unified Definition of Social XR

Towards a Unified Definition of Social XR

引用

2025 IEEE Conference on Virtual Reality and 3D User Interfaces Abstracts and Workshops, VRW 2025

作者： Moslavac, Mirta Vlahović, Sara Skorin-Kapov, Lea University of Zagreb Faculty of Electrical Engineering and Computing Zagreb Croatia

ISBN: (纸本)9798331514846

Social Extended Reality (Social XR) is an evolving domain within Extended Reality (XR), encompassing immersive technologies such as Virtual Reality (VR), Augmented Reality (AR), and Mixed Reality (MR). Despite its growing significance in both research and practice, the term 'Social XR' remains ambiguously defined, with inconsistent usage across academic and industry contexts. This paper seeks to address this challenge by synthesising existing definitions and interpretations of Social XR, aiming to encourage a clearer, unified conceptualisation. Through an analysis of current research, we identify key themes underpinning the concept of Social XR, such as shared immersive 3D environments, blending of virtual and physical elements, real-time multiuser interaction, and user representation. Thus, we propose a comprehensive definition of Social XR, aiming to stimulate further discussion, encouraging refinement toward a coherent understanding of the term. Additionally, we build on previous work and introduce the XR3C Model, for a better understanding of collaboration within immersive environments and its relation to Social XR. © 2025 IEEE.

关键词： collaboration Extended reality (XR) immersive experience multimedia communication social computing

来源：评论

学校读者我要写书评

暂无评论

Tree search hyper-heuristic with application to combinatorial optimization

引用

Journal of Heuristics 2025年第2期31卷 1-42页

作者： Gil-Gala, Francisco Javier Ɖurasevic, Marko Sierra, Maria R. Varela, Ramiro Department of Computer Science University of Oviedo Gijón 33213 Spain Faculty of Electrical Engineering and Computing University of Zagreb Zagreb 10000 Croatia

In this study, we investigate using the state space search paradigm to construct heuristics in the form of Priority Rules for combinatorial optimisation problems. This is an alternative to Genetic Programming (GP) and other hyper–heuristics, which represent the most common approach currently used. To do that, we define the problem of designing heuristics as a Constraint Satisfaction Problem and then exploit Any-Time Depth-First Search to solve it. To limit the effective size of the search space, we introduced a set of powerful pruning mechanisms, some embedded into the problem definition as constraints, while others by means of constraint propagation procedures. To further reduce the search space, we propose a heuristic procedure that allows the algorithm to discard some non-promising PRs, at low computational cost. The proposed approach, termed Systematic Search and Heuristic Evaluation (SSHE), was evaluated on two hard combinatorial optimisation problems, namely the One Machine Scheduling Problem with time-varying capacity (denoted by (1,Cap(t)||∑Tj)) and the classic Travelling Salesman Problem. The results of the experimental study show that SSHE is quite competitive with GP in building PRs;in particular, the PRs obtained by SSHE and GP showcase similar performance, but the ones produced by SSHE have generally lower size and so better readability than the PRs produced by GP. © The Author(s) 2025.

关键词： Heuristic search Hyper–heuristics One machine scheduling problem Priority rules Travelling salesman problem

来源：评论

学校读者我要写书评

暂无评论

Preference Definition in a Multi-agent System Approach for User-Centric Internet of Things Smart Environments 18th

Preference Definition in a Multi-agent System Approach for...

引用

18th International KES Conference on Agents and Multi-Agent Systems: Technologies and Applications, KES-AMSTA 2024

作者： Mandaric, Katarina Jezic, Gordan Faculty of Electrical Engineering and Computing Internet of Things Laboratory University of Zagreb Unska 3 HR Zagreb10000 Croatia

ISBN: (纸本)9789819764686

The term Ambient Intelligence (AmI) has been a topic of interest for many years, as it has the potential to transform services and environments—from living and working spaces to hospitals. Context awareness is a prerequisite for the development of truly intelligent services, i.e. it is necessary to detect and understand the state of the space as well as events/activities taking place. Additionally, for the complete adaptation of the space to the user, it is necessary to take the user’s preferences into account. Defining user preferences is a very complex area with numerous approaches, starting with the development of simple automated services in a smart home, where the user is offered control of a simple set of rules to define their wants and needs. The paper conducts an analysis of preference definitions in different environments and proposes a preference definition for the use case of smart lighting and smart heating and cooling system in a Multi-Agent System approach. This proposed definition forms the basis for effective negotiation between users with different preferences, where conflicts are inevitable. © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2025.

关键词： Smart homes

来源：评论

学校读者我要写书评

暂无评论

Community management in informal e-learning environment - analysis and vision of Facebook in Central Library of faculty of electrical engineering and computing

Community management in informal e-learning environment - an...

引用

Proceedings of the International Convention MIPRO

作者： Jadranka Lisek Dina Vrkić Ivan Jelić Central Library of Faculty of Electrical Engineering and Computing Faculty of Electrical Engineering and Computing University of Zagreb Zagreb Croatia

At the beginning of the academic year 2011-2012 the Central Library of faculty of electrical engineering and computing (FER) set foot in the world of Web 2.0 expanding to social networks: Facebook, Twitter, YouTube and Google+. The necessity to integrate social networks into library management occurred due to the presence of a large number of patrons on the Web. It also turned out that besides being a learning and resource centre, the essential concept of a modern library refers to its active participation in teaching. For the purpose of this study we have isolated the Central Library Facebook Page, as the most significant in social networking. This paper will review and analyze the reactions of users/patrons to the content offered by the Central Library FER and the interaction over a certain time period. In the long run, it is expected that the Facebook Page will allow a form of maintaining users/patrons database, close communication, development of interactive services, breaking the conventional stereotypes of libraries and librarians, in other words, opening the door to the “living-room library” concept.

关键词： Facebook Libraries Communities Fans Electronic learning electrical engineering

来源：评论

学校读者我要写书评

暂无评论

Sizing and Implementation of Phase-Shifting Transformer for Congestion Management in Renewable-Rich Regions: A Case Study of the Croatian Power Grid

引用

IEEE Transactions on Power Systems 2025年

作者： Hrgović, Ivana Pavić, Ivica University of Zagreb Faculty of Electrical Engineering and Computing Zagreb10000 Croatia

Phase shifting transformers (PSTs) are a cost-efficient solution for controlling power flow without additional operating expenses. They can enhance power distribution in transmission systems and increase system capacity, particularly for renewable energy integration. However, existing methods for PST sizing and allocation often rely on DC power flow models, which may lack accuracy, and are typically tested on simplified networks that don't fully capture real-world conditions. Furthermore, these methods often overlook practical challenges, such as space limitations in substations. This paper presents a new methodology for PST sizing and allocation in real power networks. Unlike previous approaches, our method is based on AC power flow calculations and takes into account system requirements, space constraints, and economic factors. The goal is to determine the optimal type, size, phase angle regulation range, and location of PSTs to prevent system overloading under normal and N-1 system conditions. © 1969-2012 IEEE.

关键词： congestion management phase-shifting transformer PSDF

来源：评论

学校读者我要写书评

暂无评论

Usage of databases and the information literacy of students of the faculty of electrical engineering and computing

Usage of databases and the information literacy of students ...

引用

Proceedings of the International Convention MIPRO

作者： N. Jelača Faculty of Electrical Engineering and Computing University of Zagreb Zagreb Croatia

ISBN: (纸本)9781479981748

Given the area of their research, the students of the faculty of electrical engineering and computing (FER) need the latest scientific information from both domestic and foreign journals. That is why the Croatian Ministry of Science, Education and Sports has provided them with the access to various databases of scientific and technical journals. Furthermore, the faculty of electrical engineering and computing is a provider of one of the most important databases for engineers - IEEE Explore. Students can access these databases either from their homes or from the faculty premises; they can also attend seminars on information retrieval. The goal of this article is presenting the research on students of the faculty of electrical engineering and computing, their usage of available databases and their familiarity with the search tools for those databases.

关键词： Databases Libraries Seminars electrical engineering IEEE Xplore Information retrieval Google

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：