Earticle

최근 미디어아트와 공연예술 분야에서 예술과 기술의 융합을 활용한 다원예술의 공연들이 늘어나고 있다. 공연 자와 관객들 사이의 실시간 커뮤니케이션이 이루어지는 공연예술은 관람객들이 전시의 형태로 이를 감상하는 미 디어아트와는 달리, 작가의 의도를 공연자의 퍼포먼스와 배경, 무대장치들을 통해 간접적으로 전달하는 형태를 취한다. 본 연구는 비언어적 커뮤니케이션의 형태로 관객들의 실시간 소통이 중요한 예술공연에서, 게임엔진과 실시간 인터렉션 기술들을 접목한 공연사례들을 분석해보자 한다. 게임과 영상 분야에서 주로 사용되었던 언리 얼 엔진과 비주얼 인터렉션 기술들은 실시간 리얼타임 비주얼 출력이라는 강력한 이점으로 인해 다양한 분야에 서 폭넓게 활용되고 있고, 예술공연 분야로도 점차 확대되고 있다. 따라서 미디어아트 공연분야에서 실시간 인터 렉션 기술이 접목된 다원예술 공연들을 살펴보고 어떠한 시도와 움직임들이 있는지 살펴보고 분석해보고자 한 다. 이를 통해 미디어아트 분야 뿐만 아니라, 공연예술 및 다원예술 분야에서도 미디어아트와 실시간 인터렉션 공연을 접목시킨 다양한 형태의 예술공연들이 늘어나고 관련 제작 및 방법론 연구에 도움이 되고자 한다.

The increasing prevalence of the convergence of art and technology in the field of media art and performing arts has led to a rise in the number of performances that incorporate real-time interaction technologies, such as game engines and visual interaction technologies. This study aims to analyze and examine the integration of these technologies in multi-art performances, focusing on the importance of non-verbal communication and real-time audience engagement. By exploring and analyzing the attempts and movements made in these performances, this research seeks to expand the scope of art performances that combine media art and real-time interaction, not only in media art but also in the broader fields of performing arts and multiple arts.

본 연구는 생성적 AI와 언리얼 엔진의 MetaHuman 기술을 통합하여 고급 실시간 디지털 휴먼 시스템을 구현하는 것을 탐구한다. Inworld AI의 NPC 시스템 및 NVIDIA의 Avatar Cloud Engine과 같은 기존 기술 통합 사례를 분석하여 연구에 참고하였으며, MetaHuman의 시각적 사실성과 생성적 AI의 동적 행동을 결합한 새로운 아키텍처를 제안한다. 연구 방법론은 캐릭터 제작, AI 모델 통합, 실시간 성능 최적화를 다루고 있다. 시스템은 렌더링 품질, AI 응답 정확 도, 사용자 경험을 기준으로 평가되었으며, 그 결과 이러한 통합이 보다 반응성 높고 사실적인 디지털 휴먼을 구현할 잠재력을 입증했다. 본 연구는 게임, 교육, 가상 비서 분야에서의 시사점 을 논의하고, 현재의 한계점과 향후 연구 방향을 제시한다. 이를 통해 첨단 AI 기능과 최첨단 실시간 렌더링을 결합한 프레임워크를 제공하여 지능형 가상 캐릭터 개발을 한 단계 발전시키 는 데 기여하고자 한다.

This study explores the integration of generative AI with Unreal Engine's MetaHuman technology to create an advanced real-time digital human system. We analyze existing implementations combining these technologies, including Inworld AI's NPC system and NVIDIA's Avatar Cloud Engine, to inform our research. We propose a novel architecture that leverages MetaHuman's visual fidelity and generative AI's dynamic behavior. Our methodology covers character creation, AI model integration, and optimization for real-time performance. The system is evaluated based on rendering quality, AI response accuracy, and user experience. Results demonstrate the potential of this integration for creating more responsive and lifelike digital humans. We discuss implications for gaming, education, and virtual assistants, addressing current limitations and proposing future research directions. This study contributes to the field by providing a framework for combining advanced AI capabilities with state-of-the-art real-time rendering, advancing the development of intelligent virtual characters.

방대한 빅데이터를 기반으로 발전한 인공지능 기술은 창작과 영상 예술, 인간의 고유 영역으로 여겨졌던 분야에 큰 영향을 미치기 시작했다. 전통적으로 선형적이었던 3D 애니메이션 제작 방식은 AI 이미지 생성 모델을 활용해 초기 기획, 컨셉 및 아이디어를 빠르게 비주얼화하고, 실시간 게임 엔진을 통해 실시간으로 수정 및 개선 작업을 반복하며 완성도를 높이는 비선형 적 방식으로 변화하고 있다. 본 논문에서는 Stable Diffusion의 노드 기반 비주얼 스크립트인 ComfyUI를 활용한 사전 제작 과정과 iClone 및 언리얼 엔진을 통해 디지털 휴먼을 제작하고 구체화하는 일련의 파이프라인을 분석한다. 이를 통해 전통적인 파이프라인을 넘어서는 새로운 예술적 표현의 가능성을 모색한다.

Artificial intelligence (AI) technology is increasingly influencing creative processes, video arts, and other domains previously regarded as uniquely human. Traditionally, 3D animation production followed a linear process; however, it is now transitioning toward a nonlinear approach that leverages AI image generation models. This shift enables the rapid visualization of initial plans, concepts, and ideas while enhancing the overall quality through iterative real-time modifications and improvements facilitated by real-time game engines. In this paper, we analyze the pre-production process utilizing ComfyUI, a node-based visual scripting tool for Stable Diffusion, and examine a pipeline for creating and refining digital humans using iClone and Unreal Engine. This study seeks to explore the potential for new forms of artistic expression that transcend traditional production pipelines.

K-POP의 한류가 세계적으로 주목받는 현시점에 가상공간에서의 DMX(Digital Multiplexing) 디자인은 새로운 프 리비즈(Previs)로 주목받을 것으로 사료 된다. 따라서 언리얼 엔진을 활용하여 다음과 같이 3단계로 개발하였다. 첫 번째는 자체 개발한 절차적 생성 시스템(PCG; Procedural Content Generation)으로 DMX 모듈러를 개발하였고, 두 번째 DMX 스테이지 모듈러시스템으로 소형, 중형, 문화재형, 대형, 아레나형 스테이지로 구분하여 기획, 개발하였다. DMX 스테이지를 어디든지(위치), 어떠한 사이즈(스케일)이거나, 어떠한 분위기(연출)든지, 모두 대응할 수 있는 연출시스템을 완성함으로써 연 구를 구체화 시켰다. 연구 한계점으로는 실제 관련 업체와 협력하여 연구하지 못한 점과 기술 노출을 꺼리는 기업문화의 배경 때문에 더 깊게 연계하지 못한 부분을 들 수 있다. 향후 연구로는 점점 발전하는 한류를 가상공간에서도 멋진 연출을 통해 DMX를 구현함으로써 다가올 미래 콘텐츠로서 자리매김 되기를 바라는 바이다.

With the global K-pop wave gaining global attention, DMX (Digital Multiplexing) design in virtual spaces is expected to gain traction as a new previs. Therefore, we developed this design in three stages using Unreal Engine. First, we developed a DMX modular system using our proprietary Procedural Content Generation (PCG) system. Second, we planned and developed a DMX stage modular system, categorizing it into small, medium, Korean, large, and arena stages. This research was substantiated by the completion of a production system that can accommodate any DMX stage location, size, scale, or atmosphere. Limitations of this research include the inability to collaborate with relevant companies and the reluctance of corporate cultures to disclose their technology, preventing deeper integration. Future research will explore how to leverage the ever-evolving Korean Wave through compelling productions in virtual spaces, thereby establishing DMX as a future content.

This study will concentrate on developing an automated process for creating a 3D environment utilizing satellite imagery, a segmentation algorithm, and geospatial data. Traditional methods for crafting a 3D environment primarily rely on manually sculpting terrain and generating 3D objects, which requires substantial time, effort, and resources from the developer. We aim to introduce a system that combines satellite images, digital terrain models, and building segmentation through Python programming to create 3D environments in Unreal Engine. The implementation includes a Python Tkinter GUI for data collection and preprocessing, Mask-RCNN for building segmentation, and the use of Open Street Map (OSM) data to utilize data availability and visualization of data. The system will be evaluated by generating 3D scene environments using satellite image input and incorporating geospatial datasets to analyze and measure the visual similarities between actual and generated 3D environments.

코로나 사태 이후 공연예술계는 ‘초지능화’와 ‘초연결성’의 특징으로 대표되는 메타버스에 주목 하기 시작했다. 직접 만나서 소통할 수 없는 비대면적 상황은 현실세계와 가상공간을 연결해 주는 메타버스 플랫폼 전시 공연 활성화에 큰 영향을 끼쳤다. 코로나 사태가 끝난 이후에도 공연예술계는 다양하고 실험적인 온-오프라인 공연의 형태로 해당 기술을 활용한 다양한 프로 젝트 전시들을 선보이고 있다. 이처럼 실감 콘텐츠와 가상현실을 접목한 VR, XR 기술들은 빠 르게 실감 콘텐츠와 전시공연분야에 접목되며 실험적이고 다양한 공연과 전시를 선보이고 있 다. 메타버스 플랫폼을 활용한 전시와 공연들이 관객들이 어떠한 방식으로 다가가고 있는지 실감 콘텐츠 공연사례들을 분석하고, 가상현실이 일상화되고 있는 요즘, 전시와 공연 예술 분 야의 변화의 방향을 살펴본다.

After the Corona incident, the performing arts world began to pay attention to the metaverse, which is characterized by ‘hyperintelligence’and ‘hyperconnectivity’. The non-face-to-face situation where one cannot meet in person and communicate has had a great impact on the revitalization of exhibitions and performances on the Metaverse platform, which connects the real world and virtual space. Even after the coronavirus crisis is over, the performing arts world is presenting various project exhibitions using the technology in the form of diverse and experimental online and offline performances. In this way, VR and XR technologies that combine realistic content and virtual reality are quickly being applied to the field of realistic content and exhibitions, presenting experimental and diverse performances and exhibitions. We analyze cases of realistic content performances to see how exhibitions and performances using the Metaverse platform are approached by audiences, and look at the direction of change in the field of exhibitions and performing arts these days, when virtual reality is becoming a daily routine.

본 연구는 가상공간에서 현실과 같은 활동이 활발히 이루어지고 있는 최근의 사회 현상에 관심을 두고 딱지본의 2D 이미지를 사용하여 3D 가상공간을 디자인한 작품에 관한 글이다. 가상공간은 컴퓨터의 발달과 사회, 문화적 요구에 의해 그 개념과 범위 가 점차 확대되어 상상하는 모든 형태의 현실을 만들어낼 수 있는 공간이라고 할 수 있다. 딱지본은 표지에 흥미로운 내용이 그려진 마치 아이들 놀이도구와 같이 울긋불 긋하게 인쇄된 이야기책이다. 딱지본 표지 이미지를 활용하여 과거 낭만시대로의 여 행을 표현한 3D 가상공간을 디자인하였다. 작품은 구글 지도를 캡쳐하여 단계적 변환 을 거친 3D 공간 이미지와 포토샵 편집과정을 거친 2D 딱지본 이미지들로 구성된다. 편집과 변환과정을 거친 구성요소들은 언리얼 엔진을 기반으로 화면에 배치하고 조합 하여 가상공간으로 제작하게 된다. 작품뿐만 아니라 다양한 장르에서 딱지본의 콘텐 츠와 3D 가상공간의 융합을 응용한 새로운 표현 방식과 창의적인 아이디어가 파생되기를 기대한다.

This study is about an artwork that designs a 3D virtual space using 2D images of Ttakjibon with interest in the recent social phenomenon in which realistic activities are actively carried out in virtual space. The virtual space can be said to be a space that can create any form of reality imaginable by gradually expanding its concept and scope due to the development of computers and social and cultual demands. Ttakjibon is a storybook printed in a colorful way like children’s play equipment with interesting contents on the cover. A 3D virtual space was designed to express travel to the romantic era in the past by using the images of Ttakjibon. The artwork consists of a 3D space image that has been converted by capturing Google Maps and 2D Ttakjibon images that have been edited by Photoshop. The components, edited and converted, are placed on the screen based on the Unreal Engine to create a virtual space. It is expected that new expression methods and creative ideas will be derived by applying the fusion of Ttakjibon’s contents and 3D virtual space in various genres as well as artworks.

Physics game engine, the graphics engine and network engine, artificial intelligence engine, sound engine, isometric engine that is more variety such thoughts. The engine for the convenience of the various parts during manufacture of the game. If there is no engine, it is to use this to be produced by programming a portion corresponding thereto directly, saving time and money. For example, look at some first example I saw a lot of Havok physics engine in the engine is relatively inde For example, depending on the intensity and direction of the character is pushed when the hammer hits the hammer Distance, angle, an engine that can help you easily apply a variety of calculations, such as physical strength. The following is a typical wah, etc. also wrote the Gamebryo graphics engine as. Sikimyeo optimize the graphics engine built object in 2D or 3D image through techniques such as Shader serves to make faster, more glamorous images that appear on the screen. Artificial intelligence engine to assist so that you can easily AI characters Any engine helps but, for example, A * pathfinding algorithm, easy to like Adjust the AI monster any play a role. Thus the role that can help you more easily and quickly to play the game engine is produced. In the paper, we analyze the "Unreal Engine" and "Unity3D" engine.

현재 기술이 발전하게 됨에 따라 멀티미디어 콘텐츠 제작에는 많은 비용과 인적자원이 필요하고, 수많은 멀 티미디어 제작 도구 소프트웨어가 필요하게 되었다. 이런 각각의 멀티미디어 제작 소프트웨어들은 배울 내용 이 너무 많은데다가 고품질의 최종결과물을 만들기 위해서 많은 시간이 소요되는 문제점이 있다. 그렇기 때 문에 대학생들이 실무 업계에서 통용될 수 있는 멀티미디어 콘텐츠 제작을 위해서 4년이라는 대학교육 시간 내에서 이것들을 전부 학습하기엔 무리가 있고 여러 가지 제약이 수반될 수밖에 없다. 이러한 문제점들이 대 두된 가운데 게임을 제작하기 위한 게임 엔진이 새로운 대안으로 제시되고 있다. 여러 가지 소프트웨어로 만 들어야 했던 멀티미디어 콘텐츠를 게임 엔진으로 하나로 통합하여 만들 수 있게 되었고, 실시간 렌더링을 할 수 있는 게임 엔진의 특성상 최종 결과물을 만드는 데 기존의 방식보다 비약적인 시간 단축이 가능하기 때 문이다. 본 논문에서는 대학에서 효율적인 교육을 위하여 상용게임 엔진 중 가장 많이 쓰는 Unity 엔진과 Unreal 엔진을 중심으로 각 게임 엔진의 특성과 장·단점을 비교 분석함으로써 위 두 엔진 중 대학교에서 어 떤 것을 선택하여 교육하는 것이 효율적인지에 대한 분석 결과를 제시하고자 한다.

With the development of current technology, multimedia content production requires a lot of cost and human resources, and a lot of multimedia production softwares are required. There are problems that each of these multimedia production softwares have too much to learn and takes a lot of time to produce high-quality final results. Therefore, it is difficult for college students to learn all of these within four years of college education to produce multimedia content that can be used in the working-level industry and there are many restrictions. Amid these problems, a game engine to produce games is being proposed as a new alternative. Multimedia content, which had to be made with various software, can be integrated into one game engine, and due to the nature of the game engine that can render in real time, it is possible to reduce the time to produce final results by leaps and bounds. In this paper, we want to compare the characteristics and strengths and weaknesses of each game engine with Unity and Unreal engine which are used most often among commercial game engines, and suggest an analysis of which of the above two engines is efficient in choosing and training at the university.

This study proposes a real-time content design pipeline optimized for Unreal Engine, integrating generative AI-based image creation with AI-assisted 3D modeling tools. The pipeline aims to streamline the production of high-quality assets for real-time applications, including games and simulations. Two types of subjects were selected: a bust combining organic character features, and a stone slab characterized by planar and symmetrical structure. Multi-angle image data were first synthesized using advanced generative AI models to simulate diverse viewpoints. These were then processed using AI-enhanced photogrammetry and modeling tools to reconstruct detailed 3D meshes and extract base textures. Post-processing steps, including mesh decimation, UV unwrapping, and texture baking, were performed to ensure compatibility with Physically Based Rendering (PBR) workflows used in Unreal Engine. The final assets were successfully imported into Unreal Engine, demonstrating visual fidelity and performance suitability in a real-time environment. The study confirms the pipeline’s potential for accelerating asset development and suggests promising future directions in AI-driven digital content creation.

기존 게임 개발에서 가상세계 제작은 기준 없이 게임 요소들을 하나씩 배치하는 방식으로 이루어졌다고 판단된다. 여기에서 주목할 것은 투자 대비하여 과연 양질의 가상세계를 효율적으로 제작 되었는지에 대한 궁금증이다. 본 연구에서는 언리얼 엔진에서 모듈화라는 개념을 바탕으로 효율적 가상세계를 구축할 수 있는 방법론을 제안하고자 한다. 먼저 선행사례 들을 분석하고 모듈화를 위한 기준요소를 5가지로 추출하였다. 또한 인스턴스 제작 파이프라인의 개념을 4단계로 구분하여 제안 하였으며, 도심형 가상세계 제작을 위한 최소단위의 인스턴스 모듈을 4가지로 압축하였다. 최종적으로 최소단위의 모듈 과 기준요소를 바탕으로 구성된 도심형 가상세계를 구현하여 제시하였다. 결론적으로 이러한 효율성을 중심으로 하는 제작방 식에 대한 연구는 디자이너나 아티스트가 제작에 소비해야 했던 시간을 아이디어와 창의력에만 집중할 수 있다고 사료된다. 연구 한계점으로는 기본이 되는 최소 모듈을 도시로 한정한 점과 언리얼 엔진으로 구현함에 있어 도출한 기준요소와 제작 파 이프라인등에 대한 검증이 이루어지지 않았다는 점을 들 수 있다. 이에 보다 발전적인 모듈화 연구를 통하여 다양한 가상세계 생성 방안이 도출되기를 기대해본다.

In the development of existing games, it is judged that virtual world production was done by arranging game elements one by one. What is noteworthy here is the question of whether quality virtual worlds were efficiently produced in preparation for investment. In this study, we propose a methodology that can build an efficient virtual world based on the concept of modularization in an unreal engine. First, precedents were analyzed and five reference elements for modularization were extracted. In addition, the concept of an instance production pipeline was proposed by dividing it into four stages, and the minimum-unit instance modules for urban virtual world production were compressed into four. Finally, an urban virtual world constructed based on the minimum unit module and reference elements was implemented and presented. In conclusion, research on the production method centered on this efficiency is thought to be able to focus the time that designers or artists had to spend on production only on ideas and creativity. The limitations of the research are that the basic minimum module is limited to the city, and the derived reference elements and production pipelines have not been verified when implementing them with an unreal engine. Therefore, it is expected that various virtual world creation plans will be derived through more advanced modular research.

승강기는 2만여 개의 다양한 부품들이 조합된 기계 전자적 장치로 서로 간의 밀접한 연관관계를 가지고 조직 적으로 운영되고 있다. 이러한 복잡한 승강기 구조를 직관적으로 이해하고, 운영되는 모델을 효율적으로 인식함으 로써 승강기 시스템에 대한 이해와 유지보수관리의 효율을 증가시킬 수 있을 것이다. 기존 승강기 관리 시스템은 승강기에서 발생하는 데이터를 기반으로 정보를 수집하고 이해하는 과정으로 승강기 유지 관리 시 실세계 정보의 표현 및 관리의 효율성이 부족한 구조를 가진다. 이에 본 논문에서는 승강기 실제 모델을 근간으로 효율적으로 운영할 수 있는 시뮬레이터 프레임워크을 제시하고자 한다. 구체적 사물에 대한 인식을 3D기반의 서비스 모델을 통해 구성하고, 동작 과정을 시각화함으로써 승강기 운영에 필요한 구조 및 운영방식에 대한 이해를 높일 수 있 을 것이다. 이를 위해 승강기 시스템에 대한 핵심 구성요소를 파악하고 이들 간의 관계와 운영 방식을 시각화하 고 시뮬레이터를 구현한다. 이를 바탕으로 가상공간과 현장의 효율적인 관리 구조를 통해 현실적인 정보관리 및 운영환경을 제공할 수 있을 것이다.

An elevator is a mechanical and electronic device composed of about 20,000 various parts, and is systematically operated with a close relationship with each other. By intuitively understanding this complex elevator structure and efficiently recognizing the operating model, it will be possible to increase the understanding of the elevator system and the efficiency of maintenance. The existing elevator management system is a process of collecting and understanding information based on data generated from elevators, and has a structure that lacks efficiency in expressing and managing real-world information during elevator maintenance. Therefore, this paper proposes a simulator framework that can operate efficiently based on the actual elevator model. By constructing the recognition of specific objects through a 3D-based service model and visualizing the operation process, it will be possible to enhance the understanding of the structure and operation required for elevator operation. To this end, the core components of the elevator system are identified, the relationship between them and the operating method are visualized, and a simulator is implemented. Based on this, it will be possible to provide a realistic information management and operating environment in virtual space and real platform.

Research on the visual narrative impact of Unreal Engine's Lumen lighting system in games aims to delve into how Lumen's lighting technology plays a crucial role in game design and gameplay experience, thereby enhancing the visual storytelling of games. Lumen, Unreal Engine's dynamic global illumination solution, calculates lighting and shadows in real-time during gameplay, creating a more realistic and immersive environment. Analysis indicates that Lumen technology not only provides visually realistic and dynamic lighting effects but also significantly enriches the expressiveness and immersion of the game narrative through its changes in light and shadow.

The platform for displaying cars has been changing with the times. From the popularity of paper media to the rise of computer graphics, the improvement of technology has brought more space and possibilities to the automotive industry. Yiming Zhong proposed the workflow of car configurator through Unreal Engine 4 to implement the basic functions of configuration in 2021, according to Yiming Zhong's final presentation, there is still room to improve the realism of graphics and functionality of the car configurator. Therefore, in this paper we propose to upgrade the car shaders and lighting environments according to the real-world physics and add multi-scenes switching function to car configurator. However the multi-scenes switching function also brings a large amount of data, which leads to the problem of display lag. At the end of the paper, we use the level of details(LOD) process to reduce the amount of data for real-time computing in Unreal Engine 4 and the increase of frames per second(FPS) values verifies the feasibility of our optimization solution.

Recently, in the Game industries, they are increasing to use of game engines to reduce the development cost of 3D content and software. In particular, Unreal Engine provides a blueprint visual scripting function that enables software production without programming (coding). Although High-end video content can be produced, the problem is that content development is complicated and requires advanced manpower. To solve this problem, we propose an optimized VR game context process. This is because 1) a Blueprint visual script is used, 2) VR games with various interactions can be produced, 3) Non-majors in the software field (or groups) can develop advanced content. In various related industries such as defense, medical care, manufacturing, and construction, we may easily develop any game content without programming with our refined VR rhythm action game development process. We expect to reduce the development cost with the process advantages in the game industries.

With the rapid advancement of generative artificial intelligence and virtual human technologies, AI-driven educational systems had demonstrated significant potential in enabling personalized instruction and intelligent interaction. However, existing systems commonly suffer from limited emotional expressiveness, inadequate knowledge response mechanisms, and weak multimodal integration. This study proposed a multimodal AI tutor system based on Unreal Engine, integrating key technologies such as the GPT-4 language model, VITS-based speech synthesis, and NVIDIA Audio2Face for facial animation. To enhance content accuracy and adaptive responsiveness, a dual knowledge graph framework was introduced, comprising a structured teaching knowledge graph and a student cognitive intent graph. The system employs MetaHuman for high-fidelity avatar modeling and leverages Live Link to establish synchronized speech-expression feedback. Experimental results validate the feasibility of the proposed AI tutor system in virtual education environments, providing a practical foundation for the development of future intelligent educational platforms.

Digital art creation has become an indispensable part of today's society, but traditional digital art production methods have been difficult to meet the growing creative needs of artists. Therefore, this study takes the cloud series works of artist Morten Lasskogen as an example and explores the application value of 3D Max and Unreal Engine in digital art created by analyzing the lighting effects in the works of art. This research aims to form reference materials through actual case analysis and provide artists with more efficient ideas for digital art creation.

본 연구는 기존 버추얼 프로덕션(Virtual Production, VP)이 갖는 높은 장비 의존성(LED 기반 VP)과 실제와 가상의 조명과 그림자 불일치, 스필(녹색 반사) 현상 등으로 인한 합성 품질 저하 문제(그린 크로마키 기반 VP)를 지적하고, 이를 극복하기 위한 인공지능(AI) 기반 VP 워크플로우의 가능성을 탐색하였다. 동일한 실사 푸티지를 활 용하여 그린스크린 기반 VP 방식과 AI 기반 방식을 비교 분석하였으며, 전경 분리, 리라이팅, 그림자 합성의 각 단 계를 SwitchLight와 Unreal Engine을 활용해 실험적으로 구현하였다. 실험 결과 AI 기반 방식은 라이팅의 조화와 제작 효율성 면에서 기존 방식에 비해 유의미한 개선을 보여주었다. 특히, AI 기술을 통한 전경 분리와 PBR 기반 리라이팅은 보다 정밀한 조명 재현과 빠른 시간 내 결과물 생성을 가능하게 하였다. 다만, 리라이팅은 아직 실시간 합성은 아니며, 고해상도 소스에 대한 의존성과 일부 프레임에서 발생하는 깜빡임 현상 등 아직 해결해야 할 기술적 한계도 확인되었다. 본 연구는 중소규모 제작 환경과 교육 현장에 적용할 수 있는 실용적인 AI 기반 VP 모델을 제 시하며, 향후 프레임 단위 temporal inconsistency(시공간 일관성)강화와 품질 평가 체계 구축을 통해 AI 기반 VP의 실용성과 완성도를 더욱 높이고자 한다.

This study addresses limitations of traditional Virtual Production (VP), such as high equipment dependency (LED-based VP), lighting and shadow mismatches, and spill effects in green screen VP. It explores an AI-based VP workflow as a solution. Using the same footage, green screen and AI-based methods were compared, with foreground separation, relighting, and shadow compositing implemented via SwitchLight and Unreal Engine. The AI approach improved lighting harmony and production efficiency, enabling precise lighting and faster results. However, relighting is not yet real-time and faces issues like high-resolution dependency and frame flickering. The study proposes a practical AI-based VP model for small-scale production and education, aiming to enhance temporal consistency and quality evaluation in future work.

Due to the government's announcement of the 2025 policy mandating coding education, there is a growing need for effective coding learning methods in elementary education. However, there are few methods available that can easily help younger students understand coding. While text-based coding and visual block coding methods exist, they have limitations. To address these issues, we propose a block coding learning platform that combines virtual reality (VR) technology with gamification elements. The traditional two dimensional (2D) block coding methods have some limitations, so this platform aims to overcome these by providing an environment where learners can intuitively understand and experience coding in a three dimensional (3D) virtual space. The primary goal is to enhance immersive, learner-centered experiences and improve creative problem-solving skills and computational thinking. This study proposes an experimental approach to demonstrate the effectiveness of a learning platform that combines VR technology with block coding. Furthermore, we expect that the VR-based platform will significantly contribute to improving the quality of education and promoting self-directed learning among students.

Looking at the recent game market, classic games released in the past are being re-released with highquality visuals, and users are generally satisfied. It can be said that the realization of realistic digital actors, which was not possible in the past, is now becoming a reality. Epic Games launched the MetaHuman Creator website in September 2021, allowing anyone to easily create realistic human characters. Since then, the number of animations created using MetaHumans has been increasing. As the characters become more realistic, the movement and expression animations expected by the audience must also be convincingly realized. Until recently, traditional methods were the primary approach for producing realistic character animations. For facial animation, Epic Games introduced an improved method on the Live Link app in 2023, which provides the highest quality among mobile-based techniques. In this context, this paper compares the results of animation produced using both keyframe facial capture and mobile-based capture. After creating an emotional expression animation with four sentences, the results were compared using Unreal Engine. While the facial capture method is more natural and easier to use, the precise and exaggerated expressions possible with the keyframe method cannot be overlooked, suggesting that a hybrid approach using both methods will likely continue for the foreseeable future.