Visual Editor Research Articles

Bilinear Models of Parts and Appearances in Generative Adversarial Networks.

Recent advances in the understanding of Generative Adversarial Networks (GANs) have led to remarkable progress in visual editing and synthesis tasks, capitalizing on the rich semantics that are embedded in the latent spaces of pre-trained GANs. However, existing methods are often tailored to specific GAN architectures and are limited to either discovering global semantic directions that do not facilitate localized control, or require some form of supervision through manually provided regions or segmentation masks. In this light, we present an architecture-agnostic approach that jointly discovers factors representing spatial parts and their appearances in an entirely unsupervised fashion. These factors are obtained by applying a semi-nonnegative tensor factorization on the feature maps, which in turn enables context-aware local image editing with pixel-level control. In addition, we show that the discovered appearance factors correspond to saliency maps that localize concepts of interest, without using any labels. Experiments on a wide range of GAN architectures and datasets show that, in comparison to the state of the art, our method is far more efficient in terms of training time and, most importantly, provides much more accurate localized control.

Statewise: A Petri Net-Based Visual Editor for Specifying Robotic Systems

We present Statewise, a visual editor designed to enable developers to model and simulate complex systems using colored Petri nets in an intuitive, graphical way. Utilizing Statewise, we explore two use cases to demonstrate its capabilities. We also discuss potential enhancements to further extend its applicability in more complex scenarios.

A highly-configurable session designer for VR nursing training

Virtual Reality (VR) has proven to be a valuable tool for medical and nursing education. VR simulators are available at any time and from anywhere, and can be used with or without faculty supervision, which results in a significant optimization of time, space, and resources. In this paper we present a highly-configurable session designer for VR-based nursing education following the Standards for QUality Improvement Reporting Excellence: SQUIRE 2.0 and SQUIRE-EDU. Unlike existing platforms, we focus on letting educators quickly customize the training sessions in multiple aspects. This is achieved through the use of a visual editor, where educators can choose the environment and the elements in the session, and a node-based visual system for programming the behavior of the sessions. Thanks to end-user customization, educators can develop different variants of a session, adapt the sessions to new nursing protocols, incorporate new equipment or instruments, or accommodate custom environments. We have implemented the designer and tested it on various nursing training scenarios. The designer can be integrated into VR simulators to help educators save time in delivering highly customizable training.

TopicRefiner: Coherence-Guided Steerable LDA for Visual Topic Enhancement.

This article presents a new Human-steerable Topic Modeling (HSTM) technique. Unlike existing techniques commonly relying on matrix decomposition-based topic models, we extend LDA as the fundamental component for extracting topics. LDA's high popularity and technical characteristics, such as better topic quality and no need to cherry-pick terms to construct the document-term matrix, ensure better applicability. Our research revolves around two inherent limitations of LDA. First, the principle of LDA is complex. Its calculation process is stochastic and difficult to control. We thus give a weighting method to incorporate users' refinements into the Gibbs sampling to control LDA. Second, LDA often runs on a corpus with massive terms and documents, forming a vast search space for users to find semantically relevant or irrelevant objects. We thus design a visual editing framework based on the coherence metric, proven to be the most consistent with human perception in assessing topic quality, to guide users' interactive refinements. Cases on two open real-world datasets, participants' performance in a user study, and quantitative experiment results demonstrate the usability and effectiveness of the proposed technique.

Video and Audio Deepfake Datasets and Open Issues in Deepfake Technology: Being Ahead of the Curve

The revolutionary breakthroughs in Machine Learning (ML) and Artificial Intelligence (AI) are extensively being harnessed across a diverse range of domains, e.g., forensic science, healthcare, virtual assistants, cybersecurity, and robotics. On the flip side, they can also be exploited for negative purposes, like producing authentic-looking fake news that propagates misinformation and diminishes public trust. Deepfakes pertain to audio or visual multimedia contents that have been artificially synthesized or digitally modified through the application of deep neural networks. Deepfakes can be employed for benign purposes (e.g., refinement of face pictures for optimal magazine cover quality) or malicious intentions (e.g., superimposing faces onto explicit image/video to harm individuals producing fake audio recordings of public figures making inflammatory statements to damage their reputation). With mobile devices and user-friendly audio and visual editing tools at hand, even non-experts can effortlessly craft intricate deepfakes and digitally altered audio and facial features. This presents challenges to contemporary computer forensic tools and human examiners, including common individuals and digital forensic investigators. There is a perpetual battle between attackers armed with deepfake generators and defenders utilizing deepfake detectors. This paper first comprehensively reviews existing image, video, and audio deepfake databases with the aim of propelling next-generation deepfake detectors for enhanced accuracy, generalization, robustness, and explainability. Then, the paper delves deeply into open challenges and potential avenues for research in the audio and video deepfake generation and mitigation field. The aspiration for this article is to complement prior studies and assist newcomers, researchers, engineers, and practitioners in gaining a deeper understanding and in the development of innovative deepfake technologies.

Why the use of domain-specific modeling in airworthy software requires new methods and how these might look like? (extended version)

AbstractThe use of domain-specific modeling (DSM) in safety-critical avionics is rare, even though the ever-increasing complexity of avionics systems makes the use of DSM reasonable. DSM shows its advantage especially in capturing complex systems, data and relationships. The reason for the limited use in the (safety-critical) avionics area is mainly due to the high demands on the safety of software and systems. Everything that is to be used in flight operations and development must undergo a rigorous and complex certification process. Any data used in operations must be verified. A reduction of this effort can be achieved using qualified tools. A qualified tool can either replace or support certification activities. This article elaborates different use cases of how DSM could be used in relation to airworthy software. For those use cases, we review the effort of a certification and retrieve the major shortcomings and showstoppers of available frameworks, e.g. infeasible qualification of DSM runtimes and the unavailability of qualification artifacts. Finally, we elaborate possible ways of mitigation and show the concept and first results of a new DSM framework for airworthy applications, called DOMAINES. DOMAINES covers deterministic meta-modeling up to graph-based model transformations and verified visual editing. The concept of DOMAINES and a first functional prototype are presented that indicate that the shortcomings can be mitigated.

The simulation experiment description markup language (SED-ML): language specification for level 1 version 5.

Modern biological research is increasingly informed by computational simulation experiments, which necessitate the development of methods for annotating, archiving, sharing, and reproducing the conducted experiments. These simulations increasingly require extensive collaboration among modelers, experimentalists, and engineers. The Minimum Information About a Simulation Experiment (MIASE) guidelines outline the information needed to share simulation experiments. SED-ML is a computer-readable format for the information outlined by MIASE, created as a community project and supported by many investigators and software tools. Level 1 Version 5 of SED-ML expands the ability of modelers to define simulations in SED-ML using the Kinetic Simulation Algorithm Onotoloy (KiSAO). While it was possible in Version 4 to define a simulation entirely using KiSAO, Version 5 now allows users to define tasks, model changes, ranges, and outputs using the ontology as well. SED-ML is supported by a growing ecosystem of investigators, model languages, and software tools, including various languages for constraint-based, kinetic, qualitative, rule-based, and spatial models, and many simulation tools, visual editors, model repositories, and validators. Additional information about SED-ML is available at https://sed-ml.org/.

Zagreb Library Association social media Editorial Board

In this paper, we will present the Editorial Board of Social Networks of the Zagreb Library Association (ZLA), when it was founded, how it functions, and the challenges it faces. The article will mention which social networks the Editorial Board uses to promote ZLA, and its members, but also to attract new ones. The paper will further explain how the Editorial Board chooses content to publish and how it creates its own, original content. It will present in detail all original categories, both ongoing and finished, and a free online visual editor used by the Editorial Board.
 The second part of this paper will present the analysis of the activities on each social network where ZLA has active profiles (Facebook, Instagram, LinkedIn, Twitter (today X), YouTube, Pinterest, TikTok). These statistical data will be compared: page followers and number of posts in each year from the beginning of 2020 to the first half of the current year. It will also be analysed to what extent the number of posts and their content affects the followers’ growth on each of the social networks.
 In conclusion, the paper will argue that content is not enough, it is also important how often it is published. It will present differences in the popularity and usage of social networks, the dynamics of publishing, and the choice of format, which the Editorial Board thinks is sometimes more important than the content.
 Finally, it will show the importance of its own, original content, which sets this Editorial Board apart from the content of social networks of other associations, libraries, and book bloggers. Thanks to high-quality and inspiring content, attractive visuals for announcements, and a well-thought-out publishing schedule, the Editorial Board has reached an enviable number of followers with whom it interacts regularly

TaxonWorks in its 10th Year: What’s new, what’s next?

TaxonWorks in its 10th Year: What’s new, what’s next?

Diseño de Interfaz de Usuario Web de un Editor Visual para el Modelamiento de Infraestructura en la Nube

Cloud computing has evolved the way IT technicians manage infrastructure resources to support software applications. Leasing equipment and services from cloud infrastructure providers, such as Amazon Web Services, Microsoft Azure, etc., has replaced the traditional strategy of locally installing expensive equipment. Nowadays, IT technicians model the infrastructure they need by writing scripts, then running these scripts in the provider web platform. However, writing scripts becomes a complex task that involves interacting with the Command Line Interface and knowing the commands each provider imposes. In this context, replacing textual with visual interaction becomes a need. The purpose is to fesign the user interface (UI) of a visual web editor to model the cloud infrastructure resources. The User-Centered Design (UCD) methodology was applied to design the UI. As part of DCU, a study of the UIs of diagramming online tools was conducted to identify UI design patterns; those that expert designers use when designing visual editors.The results show that a set of 11 UI patterns for designing visual editor UIs was defined. In addition, a pattern language was built considering the relationships between patterns. By using the pattern language, the visual editor UI design was composed. As conclusions, the pattern language provided a logical way to compose the visual editor UI. In this sense, the proposed UI together with the pattern language can become a reference point for designing UIs in this domain. In future work, the effectiveness of the UI in reducing the complexity of defining the cloud infrastructure will be evaluated with user tests.

Editorial Issue 34.2

Editorial Issue 34.2

Bumpy road towards a good LaTeX visual editor at Overleaf

Bumpy road towards a good LaTeX visual editor at Overleaf

Principles of creation of visual editor of XML and JSON formats

This article considers the peculiarities of the perception of large amounts of textual information, and analyzes the needs for visual editing. The implementation of a software product that works with XML and JSON formats and provides graphical and color highlighting of the main elements of the syntax. The structure of the program was developed and divided into several modules: file reading, content analysis and display. The general algorithm of work of the program is designed. The functionality for opening the file and its further analysis in the program has been created. Visualization of the read and analyzed file is presented. The aim of the article is to display the results of research on the problem of data visualization in information processing by users, as well as highlight the results of software implementation, which provides editing, creating, saving XML and JSON files and supports color and graphical separation of program text elements. At the same time, up to 7 simultaneously open files with the number of lines in the file – up to 1000 are guaranteed.

Sustaining Vision: Competency Modeling to Understand Best Practices for Visual Editors

ABSTRACT Given the rapidly evolving media ecosystem, the need for excellence in visual storytelling and high-performance among those doing that work has never been more critical. Similarly, practitioners maintain that innovation, entrepreneurial thinking, and novel solutions have never been more needed in media organizations. To help identify best practices, competency modeling is useful to optimize and understand the performance of visual editors. Subject matter interviews with 29 experts in a wide variety of media organizations reveal five competencies: professional, management, leadership, communication, and entrepreneurial. Visual editors noted that experience, technology training, media ethics, socially-supportive communication, and advocating for visual work across departments were key components of their work. Implications for best practices, selection, and training of these specialists are discussed, as well as the entrepreneurial nature of their pivotal role. A key insight for media organizations in need of innovative and entrepreneurial activity is that the change agents they need might be found in-house: among the ranks of visual editors.

Asian American Female Composers and Digital Memory

Asian American Female Composers and Digital Memory

Simulasi dosis serap kanker payudara dengan empat sudut penyinaran menggunakan teknik radioterapi konformal tiga dimensi

Cancer treatment using radiotherapy has several techniques. This study uses a three-dimensional conformal technique with four radiation directions at angles of 0°, 60°, 180°, and 240°. This study aims to determine the absorbed dose of breast cancer tissue using a three-dimensional conformal radiotherapy technique with the addition of a multileaf collimator (MLC) using a linear accelerator radiotherapy machine. The MLC is modeled in the form of a beam with 24 rods made in duplicate which is inserted in the middle of an empty space in the form of a cylinder with a diameter of 1.3 cm. The cancer tissue was modeled as a ball with a diameter of 3 cm located in the center of the left breast. The method used is the Monte Carlo method using the MCNPX Visual Editor software. The combined absorption dose for the four radiation directions showed the largest absorption dose was in cancer tissue at 2 Gy, while the organ that received the least absorption dose was the liver at 0.0054 Gy.

Пространственное распределение естественного и антропогенного эолового рельефа на севере Западной Сибири

The authors show the differentiation of the modern aeolian landforms spatial distribution (natural and anthropogenic origin) for the northern part of the West Siberian Plain (an area of about 800 thousand km2) based on the analytical processing of Sentinel-2 satellite images (resolution 10 m/px). A deciphering technique was developed; it includes the use of unsupervised ISODATA classification algorithms, visual editing, verification of results based on high-resolution satellite images, and standard spatial analysis procedures. In total, 404,675 aeolian relief objects with a total area of 2228,18 km2 were identified. The average area of natural forms was 0,44 ha (the largest share in the Yamal district of the Yamalo-Nenets Autonomous Okrug – 1,02 %), the average area of anthropogenic forms – 2,84 ha (the largest share in the Purovsky district of the same region – 0,19 %). It was revealed that in the territories of oil and gas production, the maximum area of both natural and anthropogenic aeolian forms corresponds to the boundaries of the Urengoy field (19,14 and 34,58 km2, respectively). It was established that in the Yamal Peninsula, the modern aeolian relief is confined mainly to marine and hydroglacial deposits, while in that of Gydan Peninsula, a connection can be traced with glacial or marine ones that have experienced direct mechanical action of cover ice. It was revealed that the mentioned relief is often formed in areas with increased drainage, facing towards the direction of the prevailing winds in the spring-autumn period (terraces and the watershed of the Pur and Nadym rivers). The results of the study can be used at monitoring the reclamation of disturbed areas and the natural dynamics of vegetation as an indicator of global climate change in the north of Western Siberia.

테크놀로지를 활용한 과업 중심의 L2 이중모드 협력적 글쓰기에 대한 대학생들의 성찰

This study reports the results of an open-ended questionnaire on the experiences of Korean university students who worked on task-based L2 bimodal collaborative writing enhanced by technology. Thirty-six English major students developed a reading text for 2nd-grade middle school students. Using Microsoft TEAMS as the main platform, participants conducted the collaborative writing task using real-time video conferencing, chatting, commenting, file and image posting, and (a)synchronous feedback/revision functions. Participants 1) used multiple tools to maximize their functions, 2) interacted collaboratively in every writing stage to achieve success, and 3) evaluated the quality of collaborative writing more highly than individual writing and perceived the positive impacts of collaborative writing on L2 writing in audience awareness, paraphrasing technique, and linguistic expressions. The levels of visual editing skills and awareness of online image copyright among students varied.

From Epidemic to Pandemic Modelling

We present a methodology for systematically extending epidemic models to multilevel and multiscale spatio-temporal pandemic ones. Our approach builds on the use of coloured stochastic and continuous Petri nets facilitating the sound component-based extension of basic SIR models to include population stratification and also spatio-geographic information and travel connections, represented as graphs, resulting in robust stratified pandemic metapopulation models. The epidemic components and the spatial and stratification data are combined together in these coloured models and built in to the underlying expanded models. As a consequence this method is inherently easy to use, producing scalable and reusable models with a high degree of clarity and accessibility which can be read either in a deterministic or stochastic paradigm. Our method is supported by a publicly available platform PetriNuts; it enables the visual construction and editing of models; deterministic, stochastic and hybrid simulation as well as structural and behavioural analysis. All models are available as Supplementary Material, ensuring reproducibility. All uncoloured Petri nets can be animated within a web browser at https://www-dssz.informatik.tu-cottbus.de/DSSZ/Research/ModellingEpidemics, assisting the comprehension of those models. We aim to enable modellers and planners to construct clear and robust models by themselves.

Assemble geo-analytical questions through a Blockly-based natural language interface

Abstract. Natural language Interfaces (NLIs) have the ability to make Geographic Information Systems more accessible for interdisciplinary researchers or any inexperienced users. However, the majority of research on NLIs for GIS explored NLIs for visualization or spatial data retrieval. Research on NLIs for geo-analytical questions is still lacking. Google Blockly, an open-source JavaScript library, is frequently used for developing visual programming editors for young students learning programming languages. Students set up program functions by selecting and assembling the programming blocks and the Blockly system automatically translates the block stacks into different programming languages. Similarly, we present a Blockly-based interface that generates a question depending on the blocks the user has assembled. It can be seen that a Blockly-based interface not only naturally represents syntactic structures in geo-analytical questions but also well assists users in familiarizing the blocks and generating clear and complete questions. A comprehensive usability study is still necessary to better evaluate the interface’s performance.