• International journal of advanced smart convergence
• /
• 제12권4호
• /
• pp.142-146
• /
• 2023

We present a method for generating 3D structures and rendering objects by combining VAE (Variational Autoencoder) and GAN (Generative Adversarial Network). This approach focuses on generating and rendering 3D models with improved quality using residual learning as the learning method for the encoder. We deep stack the encoder layers to accurately reflect the features of the image and apply residual blocks to solve the problems of deep layers to improve the encoder performance. This solves the problems of gradient vanishing and exploding, which are problems when constructing a deep neural network, and creates a 3D model of improved quality. To accurately extract image features, we construct deep layers of the encoder model and apply the residual function to learning to model with more detailed information. The generated model has more detailed voxels for more accurate representation, is rendered by adding materials and lighting, and is finally converted into a mesh model. 3D models have excellent visual quality and accuracy, making them useful in various fields such as virtual reality, game development, and metaverse.

• 전자통신동향분석
• /
• 제37권3호
• /
• pp.52-63
• /
• 2022

Immersive media, also known as spatial media, has become essential with the decrease in face-to-face activities in the COVID-19 pandemic era. Teleconference, metaverse, and digital twin have been developed with high expectations as immersive media services, and the demand for hyper-realistic media is increasing. Under these circumstances, MPEG-I Immersive Media is being standardized as a technologies of navigable virtual reality, which is expected to be launched in the first half of 2024, and the Audio Group is working to standardize the immersive audio technology. Following this trend, this article introduces the trend in MPEG-I immersive audio standardization. Further, it describes the features of the immersive audio rendering technology, focusing on the structure and function of the RM0 base technology, which was chosen after evaluating all the technologies proposed in the January 2022 "MPEG Audio Meeting."

• 한국의류학회지
• /
• 제47권5호
• /
• pp.929-942
• /
• 2023

Web 3.0 enables people and machines to connect, evolve, share, and use knowledge on an unprecedented scale and in new ways, drastically improving our Internet experience. The metaverse is a collective, virtual shared space supporting all digital activities. Prompted by the rapid growth of digital art and digital fashion, this theoretical analysis explores using Jean Baudrillard's simulation concept to create unique digital art non-fungible tokens (NFTs), allowing them to express and communicate ideas like real-world art. Specifically, this study analyzes 120 digital fashion portraits of humans and animals and classifies them under three types of simulacra covering four stages of Baudrillard's simulation process. The result shows that NFT fashion artworks reflect the core features of a digital reality by connecting and transcending the boundaries of cultures, genders, and nationalities. However, in the final simulation stage (the fourth step), the simulacrum can only coexist in the virtual world as a hyperreal object (the Type III of simulacrum): an object more real than reality.

• International journal of advanced smart convergence
• /
• 제12권4호
• /
• pp.286-292
• /
• 2023

This article is set against the backdrop of the rapid development of the metaverse and artificial intelligence technologies, and aims to explore the possibility and potential impact of integrating AI technology into the traditional 3D animation production process. Through an in-depth analysis of the differences when merging traditional production processes with AI technology, it aims to summarize a new innovative workflow for 3D animation production. This new process takes full advantage of the efficiency and intelligent features of AI technology, significantly improving the efficiency of animation production and enhancing the overall quality of the animations. Furthermore, the paper delves into the creative methods and developmental implications of artificial intelligence technology in real-time rendering engines for 3D animation. It highlights the importance of these technologies in driving innovation and optimizing workflows in the field of animation production, showcasing how they provide new perspectives and possibilities for the future development of the animation industry.

• 한국의류학회지
• /
• 제48권3호
• /
• pp.397-416
• /
• 2024

This study investigated gender characteristics in contemporary virtual fashion design, focusing on avatars and genderless fashion in recent collections from Auroboros, Republiqe, Placebo Digital Fashion House, RTFKT, and Tribute. Employing content analysis within a theoretical framework of gender-related research, the study coded virtual avatars in terms of biological sex, appearance, and sociological perspectives. The results showed a preference for female-type avatars, through which androgynous aesthetics were embraced and traditional gender norms were challenged. Male-type avatars reflected experimentation with blending masculine elements, emphasizing inclusivity. Human-like avatars indicated a preference for designs that promoted inclusivity and, in the process, challenged binary classifications. The examined brands strategically capitalized on compromise, sensuality, and playfulness, thereby breaking away from traditional values to opt for more diverse styles. Genderless features combined elements from traditional men's and women's clothing, espousing sensuality and playful exaggeration. These findings signify a dynamic shift away from conventional gender standards to foster inclusivity and experimentation. They can serve as a reference for promoting creative strategies and design innovation, challenging the traditional gender perspective in the fashion industry. Implementing these strategies can lead to a more inclusive representation of fashion styles, encouraging critical thinking about gender norms.

• 정보관리학회지
• /
• 제39권1호
• /
• pp.219-256
• /
• 2022

지능정보사회에서 VR 기술은 차세대 기술로서 주목받으며 그 중요성이 더욱 강조되고 있다. 이러한 배경에서 도서관에서도 VR 기술을 접목해야 할 필요성이 대두되고 있다. 본 연구는 국내외 도서관에서 이용되고 있는 VR 콘텐츠를 조사하여 분류하고 유형과 특징을 분석하는 데 목적이 있다. 따라서 본 연구에서는 VR 기술이 적용되고 있는 국내외 도서관을 대상으로 VR 콘텐츠 사례를 수집하여 분석하였으며, 분석 결과가 시사하는 바를 바탕으로 향후 VR 콘텐츠를 도서관에 적용하고자 할 때 유의해야 할 사항들을 제언하였다. 본 연구는 기존의 연구에서 중점적으로 논의되지 않았던 VR 콘텐츠를 대상으로 실제 사례에 기반하여 연구를 진행하였다는 점에서 의의가 있다.

• 한국컴퓨터그래픽스학회논문지
• /
• 제30권3호
• /
• pp.159-169
• /
• 2024

본 논문에서는 증강 현실 (AR) 기술과 사물 인터넷 (IoT) 기술을 융합하여 새로운 실시간 화재 모니터링 및 대피 내비게이션 시스템을 제안한다. 제안된 시스템은 건물 내에 설치된 IoT 온도 측정 디바이스를 통해 온도 데이터를 수집하고, 이를 IoT 플랫폼을 통해 MySQL 클라우드 데이터베이스에 자동으로 전송함으로써 실시간으로 정확한 데이터를 모니터링한다. 이후, 건축 정보 모델링 (BIM)을 통해 생성된 3D 건물 모델에 실시간 IoT 데이터를 가시화하고, AR 기술을 통해 현실 세계에 모델을 표현함으로써 직관적으로 화재 발생 위치를 파악할 수 있다. 또한, Vuforia 엔진의 Device Tracking 및 Area Targets 기능을 활용하여 사용자의 실시간 위치를 파악하고, 개선된 A* 알고리즘을 통해 여러 비상구 중 최적의 대피 경로를 찾는다. 본 논문에서는 다양한 가상 화재 시나리오를 기반으로 사용자 실험 평가를 진행하여 제안된 시스템의 실용성과 빠르고 안전한 대피 효과를 입증한다.

• 지적과 국토정보
• /
• 제52권1호
• /
• pp.105-118
• /
• 2022

최근 메타버스, 스마트시티, 디지털트윈, 자율주행차, 도심항공모빌리티 등 분야에서 3D공간객체모델 관련 정보 요구는 증가될 것이다. 공간객체에 대한 3D모델 구축은 위성·항공·지상플랫폼과 같은 다양한 장비와 모델링·인공지능·영상정합 등의 기술로 가능하다. 하지만 갱신이 필요한 공간객체를 신속하게 탐지하고 DB화하는 작업은 쉽지 않다. 이 연구에서는 공간정보(도형)과 속성을 기반으로 주소코드, 층수, 건물명, 면적 등의 매칭요소를 이용하여 건물융합DB와 변화탐지건물DB를 구축 지원할 수 있고 갱신이 필요한 객체선정의 적합성을 검증할 수 있는 시스템 프로토타입을 개발하였다. 건물융합DB 구축 시 일부 건물의 경우, 공간정보와 속성의 융합불가 및 실패 사례가 발생하여, 매칭율이 약 80%로 낮게 나타났다. 이것은 특별히 시범사업지역 내 많은 건물객체에 대한 속성정보가 누락된 것에 기인하는 것으로 판단된다. 이 연구는 3D공간객체 모델의 신속한 갱신을 위한 효율적인 드론 촬영계획 수립을 지원하여 공간객체의 중복 및 불필요한 구축 등을 사전에 방지함으로써 객체 구축 절차 개선 및 비용 절감에 크게 기여할 것이다.