• Annual Conference of KIPS
• /
• 2022.11a
• /
• pp.40-42
• /
• 2022

People's focus steadily shifted toward fashion as a popular aesthetic expression as their quality of life improved. Humans are inevitably drawn to things that are more aesthetically appealing. This human proclivity has resulted in the evolution of the fashion industry over time. However, too many clothing alternatives on e-commerce platforms have created additional obstacles for clients in recognizing their suitable outfit. Thus, in this paper, we proposed a personalized Fashion Recommender system that generates recommendations for the user based on their previous purchases and history. Our model aims to generate recommendations using an image of a product given as input by the user because many times people find something that they are interested in and tend to look for products that are like that. In the system, we first reduce data dimensionality by component analysis to avoid the curse of dimensionality, and then the final suggestion is generated by neural network. To create the final suggestions, we have employed neural networks to evaluate photos from the H&M dataset and a nearest neighbor backed recommender.

• Journal of Fashion Business
• /
• v.19 no.6
• /
• pp.112-126
• /
• 2015

In social environments, hashtags have been widely adopted and have become a new form of language for users. The current study attempts to enhance our understanding of users and their motivations to use hashtags when posting fashion-related information. Specifically, this study examines whether user characteristics (fashion leadership, conspicuousness) influence their motivations to use hashtags (curation, self-presentation, information diffusion), which then leads to behavioral intentions to continue to use hashtags and recommend the same to others. An online survey was administered to test our research questions. A total of 136 consumers in their 20s, 30s, and 40s living in Korea were used for data analysis. Structural equation modeling was conducted, which revealed that fashion leadership and conspicuousness had a positive impact on users' motivations of curation, self-presentation, and information diffusion. Motivations of self-presentation and information diffusions were found to affect users' behavioral intentions while curation had no significant impact. Practical implications are presented.

• Journal of the Ergonomics Society of Korea
• /
• v.34 no.5
• /
• pp.531-540
• /
• 2015

Objective: The objective of this research is to investigate the fundamentals of human computer interaction for wearable computers and derive technology requirements. Background: A wearable computer can be worn anytime with the support of unrestricted communications and a variety of services which provide maximum capability of information use. Key challenges in developing such wearable computers are the level of comfort that users do not feel what they wear, and easy and intuitive user interface. The research presented in this paper examines user interfaces for wearable computers. Method: In this research, we have classified the wearable user interface technologies and analyzed the advantages and disadvantages from the user's point of view. Based on this analysis, we issued a user interface technology to conduct research and development for commercialization. Results: Technology requirements are drawn to make wearable computers commercialized. Conclusion: The user interface technology for wearable system must start from the understanding of the ergonomic aspects of the end user, because users wear the system on their body. Developers do not try to develop a state-of-the-art technology without the requirement analysis of the end users. If people do not use the technology, it can't survive in the market. Currently, there is no dominant wearable user interface in the world. So, this area might try a new challenge for the technology beyond the traditional interface paradigm through various approaches and attempts. Application: The findings in this study are expected to be used for designing user interface for wearable systems, such as digital clothes and fashion apparel.

• Fashion & Textile Research Journal
• /
• v.22 no.3
• /
• pp.310-322
• /
• 2020

This study investigated the classification and characteristics of augmented reality (AR) content of fashion brands. The AR contents of fashion brands were classified according to user participation space and content delivery method. Based on these types, eight case studies were conducted, along with a discussion of AR characteristics in terms of presence, interactivity, and immersion. The results showed that AR content could be divided into four types: offline visit-external information type, offline visit-internal experience type, online utilization-external information type, and online utilization-internal experience type. It was also found that there were differences in characteristics for each type of AR content. First, the offline visit-external information type requires various new content that can provide entertainment immersion to users. Second, the offline visit-internal experience type requires a powerful inducement for users to visit a specific space providing AR content and to participate in augmented environments. Third, the online utilization-external information type needs a series of AR content that can consistently incite users' curiosity about brands and products. Fourth, the online utilization-internal experience type needs effective content to improve users' shopping experience with the virtual fitting of fashion accessories, such as eyewear, hats, jewelry, and watches. Accordingly, fashion companies should create contents that can provide appropriate presence, interactivity, and immersion by AR type.

• Journal of Fashion Business
• /
• v.26 no.3
• /
• pp.33-49
• /
• 2022

In the metaverse, it is important to embellish aesthetics of user with a character called 'Avatar', a virtual representation of the user. This study provides basic data related to the fashion trend of the metaverse by studying 'Zepeto', a representative Korean platform. For empirical research, Zepeto's "Best Items" section were investigated and analyzed in the first pre-survey. Based on this, the second and main survey was conducted using a questionnaire to investigate users' style-specific preferences and purchasing behaviors for virtual fashion, comparing style preferences between virtual and real, brand preferences, and purchasing behaviors of virtual fashion. The survey found that most users were teenage girls with a high preference for pastel-toned, feminine, and cute casual styles who had a much higher interest in brands bearing idol names than in real-world luxury brands. Many responded that they felt burdened by purchasing items that had to be purchased for cash. The same can be assumed to be the reason why they preferred a suit of items that were fully coordinated rather than individual items. These results seem to reflect characteristics of teenage girls who lack cash with a high preference for idols and feminine-cute casual styles. This study suggests considerations when creating virtual fashion items. By providing basic information, more effects and developments in creating virtual fashion items that reflect consumer preferences and reactions are expected in the future.

• International Journal of Advanced Culture Technology
• /
• v.7 no.4
• /
• pp.321-326
• /
• 2019

As AI technology is recently introduced into various fields, it is being applied to the fashion field. This paper proposes a system for recommending cody clothes suitable for a user's selected clothes. The proposed system consists of user app, cody recommendation module, and server interworking of each module and managing database data. Cody recommendation system classifies clothing images into 80 categories composed of feature combinations, selects multiple representative reference images for each category, and selects 3 full body cordy images for each representative reference image. Cody images of the representative reference image were determined by analyzing the user's preference using Google survey app. The proposed algorithm classifies categories the clothing image selected by the user into a category, recognizes the most similar image among the classification category reference images, and transmits the linked cody images to the user's app. The proposed system uses the ResNet-50 model to categorize the input image and measures similarity using ORB and HOG features to select a reference image in the category. We test the proposed algorithm in the Android app, and the result shows that the recommended system runs well.

• Journal of the Korean Society of Clothing and Textiles
• /
• v.41 no.6
• /
• pp.1022-1038
• /
• 2017

Wearable smart products contributed to expand the concept of modern clothing in terms of closer communication between wearers and the surrounding environment. This study investigated the concept of wearable smart actuators and characteristics of actuating feedback. The representative product development trends over the past 10 years are examined based on the method of actuating feedback. A first keyword search from Naver and Google reused to select the final 58 products based on the second key keyword actuating feedback. The wearable smart actuator system works in a similar way to a human body system such as those of the information processing process. Actuating feedback is delivered directly to the user as the last signal of the system, which has visual, auditory and tactile feedback. It works in complex at the device to be delivered to the user. Actuator feedback is divided into three types of active and passive, depending on the user situation, the overall purpose of the product, the collected information, and the device dependency. Active actuating feedback determines and actuates the situation on behalf of the user when the user is in an emergency or a disability situation. Passive actuating feedback plays a role in environmental monitoring to help in a healthy daily life.

• Journal of the Korean Society for information Management
• /
• v.19 no.4
• /
• pp.163-183
• /
• 2002

The purposes of this study were to examine the features of web fashion magazines and to suggest an effective website model for them. Seven web fashion magazines were analyzed. The evaluating criteria included: contents, design, navigation, and services. In addition, online-questionnaires were sent to 40 end users to obtain the information regarding user behaviors on the elected web fashion magazines. The results show that a suggested model should include various aspects: 1) providing currents information. 2) consistency in design, 3) easy in navigation, and 4) searching function etc.

• Journal of Korea Artificial Intelligence Association
• /
• v.1 no.1
• /
• pp.7-10
• /
• 2023

This study proposes an AI-based shoe recommendation model based on user clothing image data to solve the problem of the global fashion industry, which is worsening due to factors such as the economic downturn. Shoes are an important part of modern fashion, and this research aims to improve user satisfaction and contribute to economic growth through a generative AI-based shoe recommendation service. By utilizing generative AI in the personalized consumer market, we show the feasibility, efficiency, and improvements through an accessible web-based implementation. In conclusion, this study provides insights to help fulfill consumer needs in the ever-changing fashion market by implementing a generative AI-based shoe recommendation model.

• Journal of Fashion Business
• /
• v.25 no.5
• /
• pp.88-98
• /
• 2021

In this study, the characteristics of digital anthropometric methods were determined with case studies. These methods were broadly classified into two categories: non-wearable and wearable. Then, these categories were further classified into four types: 3D Scanning, mobile app, smart clothing, and smart tool Among the non-wearable types, the "3D scanning" technique was based on the use of 3D hardware equipment. With this technique, the body shape was measured and the internal body information was obtained. Therefore, it is used in fields of healthcare and fitness. Among the wearable types, "Smart clothing" involves a special clothing that measures human body and a smartphone application. Both the components are linked to a fashion platform, which is based on the measured sizes that help shoppers. The "Smart tool" has the characteristic of measuring only with smart tools and smartphone applications; it does not involve the measurement of images. The common advantage of digital anthropometric methods are as follows: they reduce the time and cost of measurement by enabling self-measurement. Moreover, simple measurements are used to determine the size of anthropometry. Thereafter, it accumulates this data to track the continuous changes in size. From an industrial point of view, digital anthropometric technology should be used to increase sales. The on-demand market can be expanded as global consumers would throng the Korean fashion market. For the consumer, an avatar should be created to fit the user's size. This would provide a fun experience to the user.