• 한국정보과학회논문지:소프트웨어및응용
• /
• 제33권12호
• /
• pp.1052-1061
• /
• 2006

본 논문에서는 베이지안 네트워크를 기반으로 생성하고 평가한 가상예제를 활용하여 범주 속성 데이타에 대한 분류 성능을 향상시키는 방안을 제안한다. 가상예제를 활용하는 종래의 연구들은 주로 수치 속성 데이타를 대상으로 하였고, 대상 도메인에 특화된 지식을 활용하여 특정 학습 알고리즘의 성능을 향상시키는 것을 목표로 하였다. 본 연구에서는 도메인에 특화된 지식을 활용하는 대신 주어진 훈련 집합을 기반으로 만든 베이지안 네트워크로부터 범주 속성 가상예제를 생성하고, 그 예제가 네트워크의 조건부 우도를 증가시키는데 기여할 경우 유용한 것으로 선별한다. 이러한 생성 및 선별과정을 반복하여 적절한 크기의 가상예제 집합을 수집하여 사용한다. 범주 속성 데이타를 대상으로 한 실험 결과, 여러 가지 학습 모델의 성능이 향상됨을 확인하였다.

• 한국의류산업학회지
• /
• 제23권4호
• /
• pp.504-514
• /
• 2021

This study attempted to extract and structure the job skills required for 3D designers, which have been recently introduced to the fashion industry. The study aimed to materialize and objectify the 3D designer's job, using a focus group interview for the survey. The 3D designer has the TD task of making 3D virtual samples using the pattern files developed in Pattern CAD. Graphic design and fabric digitization are also major tasks for the 3D designer. CLO is mainly used for 3D virtual sample production, and PixPlant, Substance, Photoshop, Cinema 4D, Daz studio, and 3ds MAX are used for image and avatar editing. As per the job skills required for 3D design work, basic knowledge about patterns and sewing, skill in 3D virtual clothing technology, ability to use various software, and English skills were considered important, in that order. In particular, the need for knowledge about patterns and sewing became more important than the skill in 3D virtual clothing technology itself. To train 3D designers, it was found that not only 3D virtual clothing software, but also education on patterns and clothing construction, CAD developer's curriculum certification system, and 3D designer qualification management were required. In addition, 3D designers are recognized as an essential job in fashion vendors, and the demand for domestic brands is increasing. The biggest limitation of the 3D virtual clothing system is the perfection of the digital fabric. Also, technical improvement is needed.

• 유통과학연구
• /
• 제22권6호
• /
• pp.33-44
• /
• 2024

Purpose: This study aims to measure how information quality, system quality, security, usefulness, and ease of use of Augmented Reality (VR) and Virtual Reality (AR) influence user satisfaction, motivating intelligent travel technology developers to improve VR/AR quality to meet customer requirements. Research design, data and methodology: This study investigates users interested in travelling in Ho Chi Minh City and Nha Trang City, Vietnam. The research model was implemented using an online questionnaire and face-to-face from 405 valid samples. To evaluate the scale's reliability, the study used the software SPSS 20. Test research hypotheses and evaluate measurement and structural models. This research uses AMOS 20 software. The proposed model is firmly grounded in the Information System Success model (ISS) and the Technology Acceptance Model (TAM), providing a solid theoretical foundation for our research. Results: Results show that consumer perceptions of information quality, system quality, security, usefulness, and ease of use have a positive impact on the perceived quality of VR/AR, thereby influencing tourists' travel intention. Conclusions: The results of this research enrich the theoretical understanding of consumer behaviour toward intelligent technology products in tourism, providing management implications for manufacturers to improve the quality of tourism products and satisfy user requirements in experience before considering choosing a destination.

• 한국의상디자인학회지
• /
• 제15권2호
• /
• pp.77-89
• /
• 2013

This study investigated the formation of silhouette and hemline shape of knit flare skirts according to the properties of knit material through virtual clothing with a 3D virtual clothing system called i-Designer of Technoa, thus building a database of the property data of knit material to reduce the number of sample making steps repeated and implemented several times in the process of clothes making. The results would help to estimate a silhouette in advance, offer assistance to the development of original knit wear, and explore ways to provide basic data for the development of the knit industry of the nation. The investigator made 12 kinds of experimental clothes to the angles(width of skirt: $90^{\circ}$ and $180^{\circ}$), gauge(7G, 12G, and 15G), and grain directions(wale and bias direction) of experimental clothes for virtual clothing. The dynamic characteristics of knit skirt samples according to each gauge were measured with the KES-FB system. Draper shapes were analyzed with the sectional shape data of hemline based on i-Designer. As for the measurements of the sectional shape of hemline and the formation of silhouette, the number of nodes, the average height of node mountains and valleys, and the hemline width right and left and before and after increased at the angle of $180^{\circ}$ than $90^{\circ}$. As gauges multiplied, the number of nodes, and silhouette angle dropping. When considering grain directions, the number of nodes and silhouette index increased in the wale direction at the angle of $90^{\circ}$ with the number of nodes and silhouette angle increasing in the wale direction at the angle of $180^{\circ}$.

• Asian Journal of Atmospheric Environment
• /
• 제12권1호
• /
• pp.78-86
• /
• 2018

Filter-based sampling techniques are the conventional way to collect particulate matter, but particles collected and entangled in the filter fibers are difficult to be removed and thus not suited for the following cell- and animal-based exposure experiments. Collecting aerosol particles in powder form using a cyclone instead of a filter would be a possible way to solve this problem. We developed a hybrid virtual-impactor/cyclone high-volume fine and coarse particle sampler and assessed its performance. The developed system achieved 50% collection efficiency with components having the following aerodynamic cut-off diameters: virtual impactor, $2.4{\mu}m$; fine-particle cyclone, $0.18-0.30{\mu}m$; and coarse-particle cyclone, $0.7{\mu}m$. The virtual impactor used in our set-up had good $PM_{2.5}$ separation performance, comparable to that reported for a conventional real impactor. The newly developed sampler can collect fine and coarse particles simultaneously, in combination with exposure testing with collected fine- and coarse-particulate matter samples, should help researchers to elucidate the mechanism by which airborne particles result in adverse health effect in detail.

• 한국방송∙미디어공학회:학술대회논문집
• /
• 한국방송공학회 1996년도 Proceedings International Workshop on New Video Media Technology
• /
• pp.18-21
• /
• 1996

In the field of Virtual Reality, many studies have been reported. Especially, there are many studies on generating virtual creatures on computer systems. In this paper we propose an algorithm to automatically generate 3D fish models from 2D images which are printed in illustrated books, pictures or handwritings. At first, 2D fish images are captured by means of image scanner. Next, the fish image is separated from background and segmented to several parts such as body, anal fin, dorsal fin, ectoral fin and ventral fin using the proposed method“Active Balloon model”. After that, users choose front view model and top view model among six samples, respectively. 3D model is automatically generated from separated body, fins and the above two view models. The number of patches is decreased without any influence on the accuracy of the generated 3D model to reduce the time cost when texture mapping is applied. Finally, we can get any kinds of 3D fish models.

• 한국의류학회지
• /
• 제45권5호
• /
• pp.907-921
• /
• 2021

This study aims to predict causal relationships between experience economy, perceived flow, and continuous use intention in the mixed reality (MR) environment. A virtual fitting mirror with two modes (i.e., avatar and self-image) was selected for this study. A total of 200 samples was obtained in two sample frames: virtual fitting users in avatar mode (group 1, n = 119) and self-image mode (group 2, n = 81). The results showed that the experience economy consists of entertainment, education, esthetic, and escapism. The entertainment and esthetic experiences had positive effects on perceived flow, leading to continuous use intentions. For avatar mode, the entertainment and esthetic experiences had positive effects on continuous use intentions through the mediating effect of flow; and the education experience had a positive effect on the continuous intentions to use the MR technology. For self-image mode, the flow mediated the effect of entertainment on continuous use intentions, whereas education and escapism directly affected the intention to use the MR technology. The paper also discusses the theoretical and managerial implications of using MR technologies in fashion retailing.

• 한국실내디자인학회논문집
• /
• 제16권2호
• /
• pp.314-321
• /
• 2007

The statistical method of human sensibility ergonomics is widely used for analyzing interior design because it has standard processes and it can help to get quantifiable results. However applying this method demands repeated intense work and great time and effort is required. In this study, a tool applying Web and virtual reality techniques for statistical human sensibility ergonomic interior design analysis is proposed and the key parts of the tool including database and interface are implemented. The database contains the sensibility adjective table and the physical interior design factor table for analyzing the relationship between human sense and physical design factors. Interface of the tool is implemented using Web technologies, so testers can evaluate interior design samples via standard Web browsers. The 3D control which is an important component of the interface is also implemented. Employing the suggested tool can reduce effort and time for evaluating human sense in Interior design field.

• 한국정보과학회:학술대회논문집
• /
• 한국정보과학회 2000년도 가을 학술발표논문집 Vol.27 No.2 (2)
• /
• pp.302-304
• /
• 2000

본 논문에서는 잡음 추가와 네트웍 앙상블을 이용하는 기법으로 최근에 제안된 가상 샘플 생성 방법(VSG:Virtual Sample Generation)을 개선하는 방법을 제안하고, 이를 대표적인 앙상블학습 알고리즘인 Bagging, Boosting과 비교한다. 기존의 가상 샘플 생성 방법에 기초하여 입력 데이터의 분포를 고려하여 가상 샘플을 생성하는 방법을 제안한다. 이 방법은 입력 분포의 밀도가 높은 곳에서 가장 샘플로 인한 과소 적합을 방지하고 밀도가 낮은 곳에서 가상 샘플로 인한 과도 적합을 방지하기 위한 것이다. 본 논문은 입력 데이터의 밀도를 추정하는 새로운 과정을 정리하고 입력 분포에 따라 적합한 가상 샘플을 생성하는 방법을 고안했다. 그리고 제안하는 방법의 일반화 성능 향상을 보이기 위해 여러 가지의 합성 데이터를 사용하여 실험을 하였고 이를 Bagging, Boosting, VSG의 성능과 비교하였다.

• 유통과학연구
• /
• 제19권11호
• /
• pp.81-90
• /
• 2021

Purpose: The purpose of this research is that Metaverse platforms' UXBDI can be distinguished among Metaverse platform channel types. Metaverse platform represents a collective term signifying services that include augmented reality, lifelogging, mirror worlds, and virtual worlds. User Experience-Based Design Innovativeness (UXBDI) is characterized by novelty in product design and services that satisfy user experience. This study examined the effect of Metaverse platforms' UXBDI on user-Metaverse platform relationships. Research design, data and methodology: Metaverse platform users were selected as samples, and a marketing research institution known as a panel company conducted the survey. It used multiple regression to test the impact on platform identification and commitment based on a survey of 442 South Korean respondents. Results: The research confirmed that UXBDI sub-dimensional scales of attractiveness and interaction increased user-Metaverse platform identification and commitment. Also, a Metaverse platform identity increased user-Metaverse platform identification in virtual and mirror worlds. Conclusion: This study contributes to multiple academic fields. First, the UXBDI of Metaverse platforms appears to be a key component of ongoing user-Metaverse relationships. Second, UXBDI affects relationships differently based on the Metaverse platform type.