• 한국정보통신학회:학술대회논문집
• /
• 한국정보통신학회 2021년도 추계학술대회
• /
• pp.555-556
• /
• 2021

우리는 패션 의류 이미지의 빠르고 정확한 분류를 달성하기 위해 최적화된 동적 붕괴 학습률과 개선된 모델 구조를 가진 딥 러닝 모델을 기반으로 하는 새로운 방법을 제안한다.

• 한국의류학회지
• /
• 제46권1호
• /
• pp.165-181
• /
• 2022

The fashion industry is creating innovative business models using artificial intelligence. To efficiently utilize artificial intelligence (AI), fashion data must be classified. Until now, such data have been classified focusing only on the objective properties of fashion products. Their subjective attributes, such as fashion brand sensibilities, are holistic and heuristic intuitions created by a combination of design elements. This study aims to improve the performance of collaborative filtering in the fashion industry by extracting fashion brand sensibility using computer vision technology. The image data set of fashion brand sensibility consists of high-end fashion brand photos that share sensibilities and communicate well in fashion. About 26,000 fashion photos of 11 high-end fashion brand sensibility labels have been collected from the 16FW to 21SS runway and 50 years of US Vogue magazines beginning from 1971. We use EfficientNet-B1 to establish the main architecture and fine-tune the network with ImageNet-ILSVRC. After training fashion brand sensibilities through deep learning, the proposed model achieved an F-1 score of 74% on accuracy tests. Furthermore, as a result of comparing AI machine and human experts, the proposed model is expected to be expanded to mass fashion brands.

• 한국의류학회지
• /
• 제29권7호
• /
• pp.897-908
• /
• 2005

The purpose of this study was to search for the effective design process model f3r mass customized clothing. Therefore, this study was to propose two models of mass customized fashion design processes which were different in the customized degree and to compare their efficiencies and appropriateness with those of the existing fashion design process. The data was obtained from a survey of 150 females in their twenties and thirties living in Seoul and Gyeonggi during April in 2003. It was analysed by frequency, $X^2-test$, crosstabulation, correlation, t-test and multiple-regression. The results of survey were: Many respondents$(62.0\%)$ preferred mass customized products and mass customized design process model which suggested more choices to presumers. The mass customized design process was considered to be applicable to the present domestic clothing market. In the case of the whole respondents, color was a very important design element in mass customized design process model; because of this, the opportunity to choose colors will be essential in mass customized design process. In the case of respondents who have higher preference on mass customized products, textile(texture) was a very important design element. In the cases of both(whole respondents and respondents who have higher preference on mass customized products), style was the most important design element in fashion design process. To summarize, it proposed that to accept the mass customized clothing will be possible in this study. What is more, the guidelines to develope mass customized fashion design process model were suggested in this study.

• 한국의상디자인학회지
• /
• 제22권3호
• /
• pp.19-32
• /
• 2020

Recently, the luxury brand market has expanded from the 5060 market to the 2030 market segment due to diversification of distribution channels, an increase in income, and changes in consumption values. This change is attributed to the increase in sales of fashion accessory products, such as bags and wallets, thereby confirming the importance of the luxury brand fashion accessories market. However, there are few studies of the luxury brand fashion accessories market or brand loyalty for fashion accessories, which accounts for a large proportion of the overall luxury brand market. The purpose of this study was to develop and empirically test a theoretical framework that captures and compares the difference of impact of service quality, product attributes, and consumer satisfaction on luxury brand loyalty between the 20-30 market and the 5060 market of the fashion accessories market. Data was collected from 598 customers who purchased luxury brand fashion accessories by objective sampling. Structural equation modelling was used to test the hypotheses developed for the study. Results found evident effects of consumer satisfaction on luxury brand loyalty through the perception of service quality by consumers. Additionally, in the 2030 market, the fashion accessory attributes had little impact on consumer satisfaction and luxury brand loyalty, whereas in the 5060 market, it had a significant impact only on consumer satisfaction, but not on luxury brand loyalty. This is the first study to examine a structural equation model of luxury brand loyalty to compare the difference of the model between segmented markets, such as the 2030 market and the 5060 market of the luxury brand fashion accessories market. These results will contribute to a better understanding of the mechanism that underlie the operation of an effective luxury branding strategy for fashion accessories.

• 한국의류산업학회지
• /
• 제7권3호
• /
• pp.291-300
• /
• 2005

The purpose of this study is to investigate, analyze the purchasing behavior of fashion goods according to life style and role model of preteen generation, and provide manager or marketing planner for the reference data so that they can understand preteen generation and make proper strategy efficiently. So called, preteen market focusing on 1014 generation (from ten to fourteen years old) is highlighted. This generation created between the year 1989 to 1993 after Seoul Olympic monopolize parent's love in abundant economic environment and rise to the core of consumption subject. Products aiming at this preteen generation continuously though consumption mind was shrunk greatly due to recession. Only 2~3 years before preteen market was regarded as grey zone which doesn't belong to not only children (between six and nine years old) but also teenagers (between fifteen to eighteen years old). But in recent day their purchasing powers have increased rapidly and age group is divided on details, so that preteen market has become a niche market. Subjects were 333 persons consisting of students in the 4th~6th grade of primary school and the 1st~2nd grade of middle school in Daegu city. Measuring instruments are as follows: 5questions to differentiate preteen generation, 22 questions to measure life style, 17questions (which have six sub-factors such as purchase motive, factor of product selection, utilization of informant, purchase time, purchase place, and purchase method) to measure the purchase behavior of fashion goods measurement, and 16 questions (which have four sub-factors such as parent, entertainer & sports stars, brothers and sisters, friends) to measure model of role. Statistical data were processed by SPSS 10.0 programs. Frequencies, Factor analysis, Cluster analysis, ANOVA, Cross analysis, Multiplex regression analysis, and Duncan's multiple range test were carried out.

• 한국의상디자인학회지
• /
• 제20권4호
• /
• pp.31-40
• /
• 2018

The purpose of this study is to identify the body type characteristics of fashion models by comparing body dimensions and body types of fashion models with ordinary women in their 20s in Korea. To that end, the study selected 71 people, with 28 being female university students in the Seoul area and 43 professional fashion models. One hundred and fifty-seven ordinary woman were selected who ranged from 20 to 29 years old, and 588 women from the 7th Size Korea fit for research purposes. Body measurement items were selected for the direct measurement data for Size Korea, which included 20 items of body size and 10 items of calculation needed for clothing production. The results of the study were as follows. According to the analysis of fashion models and ordinary woman in their 20s, their body size showed significant differences in 25 out of 30 items. The five items that do not show significant differences are bust point-bust point, waist back length, front interscye length, hip circumference-bust circumference, and neck point to bust point to waistline-waist front length. If you integrate the results, the fashion model is much taller than the ordinary woman, has a longer lower body, and has an 8.05 head figure. Also, the fashion model found itself to be the ideal type that ordinary women prefer because of the slimmer waist and bent shape, which is more than that measured in normal women.

• 패션비즈니스
• /
• 제15권3호
• /
• pp.142-155
• /
• 2011

By developing attitude transfer model, this study examined the co-marketing alliance effect between fashion and other industries (i.e., service and product brands) based upon the information integration theory. In addition, it examined the product tangibility/intangibility effects of partner brands by controlling stimuli: two alliance cases of fashion and service brands and two alliance cases of fashion and product brands. A total of 1,037 Korean women aged 20 to 39 were surveyed to compare the prior- and post- attitudes toward fashion/partner brands under four fictitious co-marketing alliance cases. Confirmatory factor analysis (CFA), multi-group CFA, structural equation modeling (SEM) analysis, and multi-group SEM analysis were conducted to test the hypotheses. The results demonstrated that the prior-attitude toward fashion brand partially affected the alliance attitude, and the co-marketing alliance was affected by prior-attitudes partner brands. The result of multi-group SEM analysis supported the significant differences between service and product brands as alliance partners, which might refer to the effect of product tangibility, existing in brand alliance contexts. The alliance evaluation affected the subsequent evaluations on each participating brands. This study empirically provides the conceptual structure of how consumer attitudes toward the participating brands interact with the attitudes toward alliance and offer practical insights. Specifically, upon employing the manipulated co-marketing alliances cases, this study demonstrates the partnering effect according to product tangibility of partner brands.

• 한국의류학회지
• /
• 제32권9호
• /
• pp.1467-1477
• /
• 2008

This research examined how customers evaluate fashion brand collaborations in speculating the collaboration factors, product, model, and designer. Respondents evaluated 8 different collaborate case cards from 1 to 8, presenting a certain fashion brand, who had recently teamed up with various kinds of product companies for collaboration. Product was found as the most important factor in evaluating the collaboration case cards, analyzed through Conjoint Analysis. Respondents segmented additionally by clothing involvement valued different collaboration factors; products for the high clothing involvement group and messenger and designer for the low clothing involvement group. All the segments and respondents chose the best the card presenting blue jeans, Mr. Chang, and B.S. Choi, as similarity for products, fame for mdoel, and similarity for designer, respectively.

• 한국의류산업학회지
• /
• 제23권3호
• /
• pp.347-357
• /
• 2021

This study analyzed the effects of optimism and innovativeness, the two drivers of a consumer's technological readiness to use an augmented reality(AR) fashion shopping service within the framework of the Technology Acceptance Model (TAM). The survey was conducted on 249 people (males = 58; females = 191) in their 20s who were willing to participate in the AR experience. Data were analyzed using confirmatory factor analysis (CFA) and structural equation modeling by AMOS 22.0. The results indicated that a greater level of perceived optimism had a positive influence on the ease of use and usefulness of the technology. The findings also show that consumers with a high degree of innovativeness tend to have a higher level of playfulness toward AR fashion shopping. Regarding the effects of user beliefs, ease of use had a positive effect on the perception of usefulness and playfulness with higher levels implying a higher consumer intention to adopt an AR fashion shopping service. In addition, this study reveals the moderating effect of consumers with high-fashion versus those with low-fashion innovativeness. For the latter, technological innovation had an insignificant effect on playfulness, thus indicating that consumers with low interest in fashion did not enjoy AR fashion shopping even if the technology was highly innovative. Nevertheless, the analysis confirms the possibility that experiencing a fashion product through AR technology could replace the actual experience of wearing the products.

• 복식문화연구
• /
• 제24권2호
• /
• pp.263-272
• /
• 2016

Using the extended technology acceptance model (TAM), the study aimed to understand consumers' adoption process for smart fashion products. The research model was designed to examine the impacts of perceived ease of use and usefulness on attitude and behavior intention toward smart fashion products based on the technology innovativeness, enjoyment, and subjective norm variables. An online survey was conducted on consumers by employing a marketing research company. A total of 230 useable responses were obtained. Confirmatory Factor Analysis (CFA) was performed to test the measurement model. The proposed hypotheses were tested by employing the Structural Equation Model (SEM). The results found a positive impact of perceived ease of use on usefulness and a positive influence of usefulness on attitude and behavior intention. Attitude had a positive effect on behavior intention. In addition, technology innovativeness was found to have a positive influence on perceived ease of use and enjoyment had a positive influence on usefulness and attitude. Subjective norm predicted behavior intention. The findings of the study contribute to smart fashion literature and have important implications for smart fashion product developers and marketers, as they offer insights into the important role of technology innovativeness, enjoyment, and subjective norms perceived by consumers in improving attitudes and behavior intentions toward the products. Limitations and future research directions are discussed.