Abstract : This study analyzes the visual communication strategies combining cinematic aesthetics and 3D graphic technology in Burberry's fashion advertising campaigns directed by the French creative group Megaforce. The research explores how advertising extends beyond a medium of product promotion to become an artistic form that creates emotional experiences. The analysis reveals that Megaforce's direction visually embodies Burberry's brand philosophy of freedom, exploration, innovation, and solidarity through cinematic composition and advanced visual effects, thereby enhancing brand value through emotional immersion. In particular, technologies such as 3D graphics, physics-based simulation, and live-action compositing serve as key tools that blur the boundary between reality and the surreal, reinforcing the brand's emotional identity. All three campaigns integrate cinematic rhythm, camera movement, and choreographic performance to deliver brand messages sensorially, presenting new possibilities for an aesthetic paradigm that merges technology and art. This study demonstrates that technology functions not merely as a visual device but as a fundamental means of communication in fashion advertising, and provides a foundation for future research on digital media-based brand communication.
Keyword : Cinematic Visuality, 3D Graphics, Fashion Advertising, Visual Communication, Brand Identity
http://dx.doi.org/10.29056/jdaem.2025.12.01
Research on Integrating Generative AI in Public Service Design
Abstract : As digital transformation accelerates, customer-centered innovation is becoming an urgent challenge in the public service sector. Against this background, generative artificial intelligence (AI) technology has been attracting attention as a key driver of new possibilities for public service design. This study proposes a new framework to accelerate service innovation by integrating generative AI technology into public service design, a human-centered problem-solving methodology, and verifies its effectiveness by applying it to a real public service design project. Based on the UK Design Council's Double Diamond model, the core framework of the research is a human-AI collaboration model that strategically incorporates generative AI throughout the four phases of the design process: Discover, Define, Develop, and Deliver. Research has shown that generative AI can be more than just an automation tool, but an ‘intelligent collaboration partner’ that complements and augments human creativity and analytical skills,data analysis to idea generation to prototyping. A prerequisite for the successful adoption of a generative AI-integrated public service design framework is the development of ethical and legal frameworks for data privacy, bias, and accountability that keep pace with technological advances.
Keyword : Generative Artificial Intelligence, Public Service Design, Double Diamond Model, Framework, Universal Design
http://dx.doi.org/10.29056/jdaem.2025.12.02
Abstract : Toothpaste is an essential oral-care product and has drawn increasing public attention. As Chinese consumers’ awareness of oral health rises, parents’ demands regarding the safety, fun elements, and flavors of children’s toothpaste have also strengthened. Although the Chinese children’s toothpaste market is currently led by foreign brands, domestic brands are gaining recognition, making competitiveness a key issue. This study analyzes the visual elements of children’s toothpaste package design in China to identify user preferences and propose design directions aligned with children’s emotional needs. The study examines the top 10 brands in the “2024 Children’s Toothpaste Brand Ranking” and analyzes color, illustration, layout, and typography using tools such as the Kobayashi Image Scale and Symbolic Affective Scale. The findings indicate that using high-contrast and vibrant colors, bold calligraphic typography, large illustrations arranged vertically, and popular tooth-related animation characters not only aligns better with children’s emotional responses but also reduces their resistance to brushing. These results demonstrate that visual elements can significantly influence brand preference.
Keyword : Package Design, Children's, Toothpaste, Visual Element
http://dx.doi.org/10.29056/jdaem.2025.12.03
Abstract : In order to solve the problems such as the lack of smooth connection between education and practice and the theory of academic barriers existing in innovative start-up education in Chinese universities, this study seeks a way to converge based on the STEM education ideology that emphasizes integration between departments. Based on the integrated theory of STEM education and innovative start-up education, the text presents a framework for innovative start-up education that integrates the concept of STEM by systematically organizing research results and research gaps based on related previous research results and empirical paths inside and outside China. In response to the current lack of empirical support and actionable measures for Chinese research, this study constructs a STEM innovative start-up education model that includes five elements and conducts empirical verification through two repetitive educational experiments. Through this, it shows that this model can significantly improve students' comprehensive literacy such as class satisfaction, knowledge integration, and practical skills. The research introduces modules such as ‘distribution of roles between disciplines’ and ‘discussion of cases of science and technology ethics’ to effectively solve the core challenges of the convergence process, prove the effectiveness and viability of the model, and provide a practical path for reference to universities.
Keyword : STEM Education, Entrepreneurship Innovation Education, Education Model, Empirical Research
http://dx.doi.org/10.29056/jdaem.2025.12.04
Abstract : This study conducted a systematic literature review of 33 research articles on chatbots in the cultural heritage field published since 2016, aiming to analyze overall research trends. The findings indicate that chatbot research in this domain has predominantly focused on museums and on providing informational services, although its scope has gradually expanded to include various heritage contexts such as historic sites and intangible heritage. Research methods were largely centered on system design, while studies grounded in user experience remained limited. In terms of target users, visitors and tourists were the primary focus, though some studies also considered learners, heritage managers, and global users. The functions of chatbots, which initially focused on information delivery, have evolved into multifunctional structures encompassing learning support, tour guidance, engagement and immersion, promotion, and preservationarchiving. Regarding technical approaches, non-generative models were dominant; however, since 2021, the emergence of generative AI-based studies has expanded possibilities for personalized and interactive experiences. Overall, chatbot research in the cultural heritage field is transitioninga technology-driven phase toward approaches that emphasize user experience. Future research should address the development of content structures tailored to heritage types and the design of user experience-centered chatbot systems.
Keyword : Cultural Heritage, Generative AI chatbots, Research Trends, Utilization, Systematic Literature Review(SLR)
http://dx.doi.org/10.29056/jdaem.2025.12.05
Performance Analysis of YOLO-Based Road Damage and Vehicle Detection
Abstract : In this study, road damage and vehicle data were collected and analyzed for the management of the Life Cycle Cost (LCC) of road facilities. Additionally, for vehicle detection, the YOLOv8n model was trained with the MIO-TCD dataset, achieving an accuracy of 82.2% based on mAP@0.5. For pothole detection, the RDD2022 dataset was utilized, and a performance comparison of some YOLO models was conducted. As a result of the performance comparison, the YOLOv5n model showed the most stable and highest mAP@0.5 (0.271). Furthermore, road damage classes were subdivided into 8 types, including crazing cracks, transverse, and longitudinal cracks, and data refinement was performed by establishing labeling criteria.The implemented system automatically calculates ESAL based on the detected vehicle data and predicts the fatigue of the road pavement, which can be utilized for establishing preventive maintenance plans. Conclusively, this study successfully derived an active traffic data collection method that deviatesconventional methods and an efficiently proven pothole detection model, which will greatly contribute to significantly enhancing the completeness of future intelligent road management systems.
Keyword : YOLO, Vehicle Detection, Road Damage, Life Cycle Cost, MIO-TCD
http://dx.doi.org/10.29056/jdaem.2025.12.06
A Study on Clinical-Context Gated Network for Precision Diagnosis of Neonatal Respiratory Diseases
Abstract : In neonatal intensive care units (NICUs), chest radiography (CXR) is a primary tool for diagnosing respiratory diseases. Previous deep-learning studies have improved diagnostic performance by simply concatenating gestational age (GA) and birth weight (BW) with image inputs via early fusion. However, this strategy suffersinformation density imbalance in the high-dimensional image space, where a few scalar clinical variables are diluted, and it does not explicitly model interactions between image features and clinical context. In this study, we propose a Clinical-Context Gated Network (CCGN) in which clinical information dynamically modulates image feature extraction. CCGN generates channel-wise attention weightsclinical variables and selectively emphasizes image features that are critical for disease discrimination. Using a large-scale multicenter dataset of 43,338 neonatal CXRs, the proposed model achieved an F1-score of 0.8728 and a macro-area under the receiver operating characteristic curve (Macro-AUROC) of 0.9457, outperforming both the ResNet50 baseline (F1-score 0.8368, Macro-AUROC 0.9122) and a previous early-fusion model (F1-score 0.8632, Macro-AUROC 0.9352). CCGN also showed consistently superior performance compared with recently proposed generic backbones such as ConvNeXt-Tiny and Swin-T. Notably, CCGN substantially improved the classification performance of transient tachypnea of the newborn (TTN), which is particularly challenging to diagnose, demonstrating its potential as a clinically useful decision-support tool in NICU practice.
Keyword : deep learning, neonatal respiratory diseases, multimodal fusion, attention mechanism, medical image analysis
http://dx.doi.org/10.29056/jdaem.2025.12.07
Abstract : This study analyzes the impact of the convergence of technology and art in the digital age on the forms and functions of street art, and explores how such technological integration has expanded the sociocultural dynamism of street art by focusing on representative domestic and international works presented at the Seoul Street Arts Festival (SSAF)2018 to 2024. Digital technologies such as Virtual Reality (VR), Artificial Intelligence (AI), Augmented Reality (AR), eXtended Reality (XR), the metaverse, interactive media, and projection mapping have diversified the expressive modes of street art while emerging as key media that actively engage audiences in the creative process. In particular, AI- and data-driven street art has evolved beyond technical experimentation to establish a new aesthetic form that fosters democratic expression and community building. As an art form situated in urban public spaces, street art distinguishes itselftraditional performing arts through its emphasis on artistry, publicness, and participation. Case analyses reveal that street art contributes meaningfully to sociocultural domains such as urban regeneration, cultural diversity, and social communication. Furthermore, it demonstrates the potential to transcend the boundaries of a marginal art form and to transform urban space into a next-generation performing arts platform by expanding physical and social boundaries through digital technologies. Building on these discussions, this study offers an in-depth examination of the evolving trajectories of street art in the digital age and its broadened sociocultural implications.
Keyword : Digital Technology, Street Art, Convergence, Urban Public Space, Next-generation Performing Arts
http://dx.doi.org/10.29056/jdaem.2025.12.08
The Effects of Short-form Content Consumption Habits on the Formation of Social Relationships
Abstract : This study empirically examined the effects of short-form content viewing on online and offline communication patterns. Using quantitative analysis, most hypotheses were not supported and several relationships were negative. In online communication, greater short-form viewing time had a statistically significant negative effect on the use of nonverbal digital expressions, while it was not significantly related to preference for immediacypreference for quantified feedback. In face-to-face (offline) communication, increased short-form viewing time was associated with reductions in concise verbal expression and in nonverbal cues such as images and gestures; no significant relationship was found with immediate empathic expression. These findings suggest that exposure to short-form content may lead not to richer communication but rather to a narrowing and simplification of expressive formats. The study’s academic contribution lies in quantitatively delineating the link between short-form media consumption and communication across both online and offline contexts, and it offers practical implications for interpersonal relationships and social interaction among young adults. Limitations include a cross-sectional design and a single-age sample; future research should include diverse age groups and content contexts.
Keyword : Short-form Content, Online Communication, Offline Communication, Media Usage Patterns
http://dx.doi.org/10.29056/jdaem.2025.12.09
A Study on the First-Person Narrative Characteristics of the Dailytoon
Abstract : This study aims to analyze the first-person narrative structure and narrative characteristics of the Dailytoon genre on Naver Webtoon, in order to examine how Dailytoons construct emotional narratives and produce social meanings. According to the Webtoon Industry Survey Report, the ‘daily life’ genre has emerged as one of the most preferred fields for future planning and production among both creators and companies, and the number of serialized works within this genre continues to grow. Based on this trend, the study selected 34 Dailytoons that were either ongoingcompleted in 2025 and conducted a qualitative textual analysis focusing on narrative perspective, content, and modes of self-representation by the authors. The analysis revealed that approximately 94% of the examined works employed a first-person point of view, wherein the author simultaneously serves as both narrator and protagonist, thereby constructing a self-representational narrative structure. The first-person narrative in Dailytoons extends beyond mere autobiographical documentation, expressing the author's emotions, occupational identity, and humor through self-reflective irony. It mediates empathy between the author and readers while transforming private experiences into shared affective narratives. Ultimately, this study identifies Dailytoons as a distinctive form of digital affective narrative that converts individual experiences into collective emotions. By revealing how the first-person narrative operates as a core mechanism of emotional authenticity and social resonance, this research provides a theoretical foundation for future studies on self-narrative structures and the industrial expansion of webtoon storytelling.
Keyword : Dailytoon, first-person point of view, first-person narrative, webtoon
http://dx.doi.org/10.29056/jdaem.2025.12.10
Abstract : This study explores the expressive effects of lines by leveraging the structural characteristics of robots, based on the concept of drawing in contemporary art. It engages with fundamental discussions on agency in creative acts mediated by the Other in traditional art, and suggests a model of human-machine collaboration that goes beyond mere convenienceefficiency by balancing the artist’s intention with the machine’s automaticity. The research defines the minimum degrees of freedom required for robotic drawing and conducts inverse kinematics analysis and simulation. An algorithm was implemented to calculate the joint angles corresponding to the brush position, pressure, direction, and tilt. Furthermore, considering the flexible properties of the brush, the study derives allowable pressure ranges based on changes in brush tilt and verifies the conditions under which variations in line thickness and stroke expression become possible. In conclusion, this research evaluates the robot not as mere tool for imitating human artworks but as a collaborative agent in the creative process.
Keyword : drawing robot, robotic art, human-robot interaction, brush stroke
http://dx.doi.org/10.29056/jdaem.2025.12.11
Abstract : This study systematically analyzed the small-scale pre-production process of converged content produced using the Unreal Engine, comprehensively exploring key issues arising during new technology adoption and corresponding resolution strategies. The case study focused on the production of Stage Cinema <Echoes of Godot>. The research method involved the researcher directly participating in the practical production environment, collecting and analyzing field observations and qualitative data. This approach addressed key pre-production issues such as the technical adaptation of actors and staff, collaboration with external partners, and the technical limitations of facial and motion capture. The research identified three key strategies for overcoming challenges in small-scale production environments: first, phased pre-testing; second, data standardization; and third, iterative experimentation and the creative acceptance of serendipity. Comprehensively, by examining how technical, artistic, and managerial capabilities organically converge during the small-scale pre-production process of Unreal Engine-based convergent content to seek realistic challenges and creative solutions, this study will provide practical production process roadmaps and reference materials for small-scale independent film and experimental content production using Unreal Engine.
Keyword : Unreal Engine, fusion content, small-scale production, digital humans, pre-production
http://dx.doi.org/10.29056/jdaem.2025.12.12
Table of Contents