• Proceedings of the Korea Information Processing Society Conference
• /
• 2022.11a
• /
• pp.432-434
• /
• 2022

우리는 전이 학습을 이용하여 원하는 특정 패션 스타일 분류기를 학습하였다. 패션 스타일 검색 결과물을 온라인 쇼핑몰과 연결하는 웹 서비스를 사용자에게 제공한다. 패션 스타일 분류기는 구글에서 이미지 검색을 통해 수집된 데이터를 이용하여 ResNet34[1]에 전이 학습하였다. 학습된 분류 모델을 이용하여 사용자 이미지로부터 패션 스타일을 17가지 클래스로 분류하였고 F1 스코어는 평균 65.5%를 얻었다. 패션 스타일 분류 결과를 네이버 쇼핑몰과 연결하여 사용자가 원하는 패션 상품을 구매할 수 있는 서비스를 제공한다.

• Journal of the Korea Convergence Society
• /
• v.11 no.12
• /
• pp.113-123
• /
• 2020

In this paper, big data analysis was conducted for past and present fashion trends and fashion cycle. We focused on daily look for ordinary people instead of the fashion professionals and fashion show. Using the social matrix tool, Textom, we performed frequency analysis, N-gram analysis, network analysis and structural equivalence analysis on the big data containing fashion trends and cycles. The results are as follows. First, this study extracted the major key words related to fashion trends for the daily look from the past(1980s, 1990s) and the present(2019 and 2020). Second, the frequence analysis and N-gram analysis showed that the fashion cycle has shorten to 30-40 years. Third, the structural equivalence analysis found the four representative clusters. The past four clusters are jean, retro codi, athleisure look, celebrity retro and the present clusters are retro, newtro, lady chic, retro futurism. Fourth, through the network analysis and N-gram analysis, it turned out that the past fashion is reproduced and evolves to the current fashion with certain reasoning.

• Journal of Korean Home Economics Education Association
• /
• v.34 no.3
• /
• pp.1-23
• /
• 2022

This study examined the perceptions of Myanmar university students and professors regarding the status and necessity of higher education programs in fashion. Data were collected from professors in textile engineering at Yangon Technological University and Myanmar university students. Closed- and open-ended questions were asked either through interviews or by email. The responses were analyzed using keyword extraction and categorization, and descriptive statistics(closed questions). Generally, the professors perceived higher education, as well as the cultural industries including art and fashion, as important for Myanmar's social and economic development. According to the students interests in pursuing a degree in textile were limited, despite the high interest in fashion. Low wages in the apparel industry and lack of fashion degrees that meet the demand of students were cited as reasons. The demand was high for educational programs in fashion product development, fashion design, pattern-making, fashion marketing, branding, management, costume history, and cultural studies. Students expected to find their future career in textiles and clothing factories. Many students wanted to be hired by global fashion brands for higher salaries and training for advanced knowledge and technical skills. They perceived advanced fashion education programs will have various positive effects on Myanmar's national economy.

• Journal of Fashion Business
• /
• v.18 no.1
• /
• pp.164-181
• /
• 2014

As the boundary between fashion and architecture is getting blurred, the interactions of the two fields are turning out abundant as well as essential. This study investigates the tectonic strategies in architectural fashion design as a novel aesthetic in the 21st century by combining literary survey and case analysis on architecture and contemporary fashion. The tectonic strategies in the works of architectural fashion designers were categorized as follows: organic geometry, technological garment construction, and independent space. Organic geometry transforms basic geometric shapes into subtle organic forms after being thrown on the body. Technological garment construction explores the garment structure and volume by applying the structural principle of suspension and fractal geometry. Independent space refers to maintaining the firm three-dimensionality of garment structure which keeps the distance from the body, assuming the similarity to architecture.

• Journal of Practical Engineering Education
• /
• v.11 no.2
• /
• pp.175-180
• /
• 2019

Existing fashion web sites show only the search results for one type of clothes in items such as tops and bottoms. As the fashion market grows, consumers are demanding a platform to find a variety of fashion information. To solve this problem, we devised the idea of linking image classification through deep learning with a website and integrating SNS functions. User uploads their own image to the web site and uses the deep learning server to identify, classify and store the image's characteristics. Users can use the stored information to search for the images in various combinations. In addition, communication between users can be actively performed through the SNS function. Through this, the plan to solve the problem of existing fashion-related sites was prepared.

• Magazine of the Korea Institute for Structural Maintenance and Inspection
• /
• v.18 no.2
• /
• pp.60-63
• /
• 2014

• Proceedings of the Korean Society of Computer Information Conference
• /
• 2020.07a
• /
• pp.559-560
• /
• 2020

우리는 패션 의류 이미지의 빠르고 정확한 분류를 달성하기 위해 최적화 된 동적 감쇠 학습률과 개선 된 모델 구조를 갖춘 딥 러닝 모델을 기반으로 하는 새로운 방법을 제안했습니다. 우리는 Fashion-MNIST 데이터 셋에 대해 제안 된 모델을 사용하여 실험을 수행하고 이를 CNN, LeNet, LSTM 및 BiLSTM의 방법과 비교했습니다.

• Journal of KIISE:Computing Practices and Letters
• /
• v.9 no.5
• /
• pp.541-550
• /
• 2003

It is important for the strategy of product sales to investigate the consumer's sensitivity and preference degree in the environment that the process of material development has been changed focusing on the consumer renter. In the present study, we propose the Fashion Design Recommender System (FDRS) of textile design applying collaborative filtering personalization technique as one of methods in the material development centered on consumer's sensibility and preferences. In collaborative filtering personalization technique based on textile, Pearson Correlation Coefficient is used to calculate similarity weights between users. We build the database founded on the sensibility adjective to develop textile designs by extracting the representative sensibility adjective from users' sensibility and preferences about textile designs. FDRS recommends textile designs to a consumer who has a similar propensity about textile. Ultimately, this paper sugeests empirical applications to verify the adequacy and the validity on this system with the development of Fashion Design Recommender System (FDRS)

• Journal of Korea Society of Industrial Information Systems
• /
• v.27 no.6
• /
• pp.13-23
• /
• 2022

In this paper, we propose a detailed component image classification algorithm by fashion item for unstructured data retrieval in the fashion field. Due to the COVID-19 environment, AI-based online shopping malls are increasing recently. However, there is a limit to accurate unstructured data search with existing keyword search and personalized style recommendations based on user surfing behavior. In this study, pre-processing using Mask R-CNN was conducted using images crawled from online shopping sites and then classified components for each fashion item through CNN. We obtain the accuaracy for collar of the shirt's as 93.28%, the pattern of the shirt as 98.10%, the 3 classese fit of the jeans as 91.73%, And, we further obtained one for the 4 classes fit of jeans as 81.59% and the color of the jeans as 93.91%. At the results for the decorated items, we also obtained the accuract of the washing of the jeans as 91.20% and the demage of jeans accuaracy as 92.96%.

• Journal of Fashion Business
• /
• v.28 no.1
• /
• pp.83-97
• /
• 2024

A plantar pressure mat with visual notifications was developed to confirm whether individuals can effectively balance themselves and correct imbalances. The sensor-embedded mat was made with a commercial yoga mat, and was tested on seven working women in their 30s to determine plantar pressure distribution when standing and squatting, and if they could recognize and correct imbalances with visual feedback. The study found that visual notifications significantly changed the plantar pressure ratio of the forefoot and hindfoot, as well as the left and right foot plantar pressure ratio. Without notifications, the center of gravity was more concentrated in the rear foot than the forefoot in both standing and squatting positions. Visual notifications showed that the center of gravity, which was largely focused on the rear foot, was distributed to the forefoot, resulting in a more evenly distributed center of gravity throughout the sole. For the change in left and right plantar pressure, the weight that was largely loaded on the left side was distributed to the right foot through the visual notification mat, confirming a more balanced plantar pressure.