• 제목/요약/키워드: Fashion AI

검색결과 76건 처리시간 0.021초

AI기반 의류정보를 이용한 비인가 접근감지 (Detection of unauthorized person using AI-based clothing information analysis)

  • 신성윤;이현창
    • 한국컴퓨터정보학회:학술대회논문집
    • /
    • 한국컴퓨터정보학회 2019년도 제60차 하계학술대회논문집 27권2호
    • /
    • pp.381-382
    • /
    • 2019
  • Recently, various search techniques using artificial intelligence techniques have been introduced. It is also possible to use the artificial intelligence to grasp customer propensity. Analyzing the clothes that customers usually wear, it is possible to analyze various colors such as favorite colors, patterns, and fashion styles. In this study, we use artificial intelligence technology to create an application that distinguish between adults and children by combining various factors such as shape, type, color and size of human clothes. Through this, it will be possible to utilize it in a living area where children can be protected in advance by grasping the intrusion of unauthorized adults in the living area where children live mainly. In addition, in the future, we can obtain good results to detect stranger adult person if we apply this experimental result to the detection system using clothing information.

  • PDF

대화질의 기반 패션 추천시스템을 위한 데이터 전처리 방법에 관한 연구 (A study on data preprocessing method for conversational query-based fashion recommendation system)

  • 최철웅;염성웅;김경백
    • 한국정보처리학회:학술대회논문집
    • /
    • 한국정보처리학회 2021년도 추계학술발표대회
    • /
    • pp.815-818
    • /
    • 2021
  • 현재 대부분의 패션 추천시스템은 프로필 또는 설문조사를 통해 수집 된 사용자의 정적 정보를 활용하고 있다. 사용자의 정적 정보는 매우 한정적이며 이를 활용하여 다양한 환경에 적합한 패션 코디셋을 추천하기란 매우 어렵다. AI코디네이터와 사용자간의 지속적인 대화가 담긴 대화질의 데이터셋을 사용하면 사용자의 상황과 환경을 고려하여 개인에게 최적화 된 패션 코디셋을 추천할 수 있다. 본 논문에서는 한국전자통신연구원(ETRI)에서 제공하는 AI 패션 코디네이터와 사용자의 대화 정보가 담긴 FASCODE 데이터셋을 사용하여 사용자의 발화에 따라 의상을 추천하는 인공지능 모델을 위한 대화질의 데이터 전처리 방법을 제안한다.

Meta's Metaverse Platform Design in the Pre-launch and Ignition Life Stage

  • Song, Minzheong
    • International Journal of Internet, Broadcasting and Communication
    • /
    • 제14권4호
    • /
    • pp.121-131
    • /
    • 2022
  • We look at the initial stage of Meta (previous Facebook)'s new metaverse platform and investigate its platform design in pre-launch and ignition life stage. From the Rocket Model (RM)'s theoretical logic, the results reveal that Meta firstly focuses on investing in key content developers by acquiring virtual reality (VR), video, music content firms and offering production support platform of the augmented reality (AR) content, 'Spark AR' last three years (2019~2021) for attracting high-potential developers and users. In terms of three matching criteria, Meta develops an Artificial Intelligence (AI) powered translation software, partners with Microsoft (MS) for cloud computing and AI, and develops an AI platform for realistic avatar, MyoSuite. In 'connect' function, Meta curates the game concept submitted by game developers, welcomes other game and SNS based metaverse apps, and expands Horizon Worlds (HW) on VR devices to PCs and mobile devices. In 'transact' function, Meta offers 'HW Creator Funding' program for metaverse, launches the first commercialized Meta Avatar Store on Meta's conventional SNS and Messaging apps by inviting all fashion creators to design and sell clothing in this store. Mata also launches an initial test of non-fungible token (NFT) display on Instagram and expands it to Facebook in the US. Lastly, regarding optimization, especially in the face of recent data privacy issues that have adversely affected corporate key performance indicators (KPIs), Meta assures not to collect any new data and to make its privacy policy easier to understand and update its terms of service more user friendly.

이미지 분석을 통한 자동화 의류 분석 시스템 (Automated Clothing Analysis System through Image Analysis)

  • 최문혁;이석준;이학재;김소영;문일영
    • 한국정보통신학회:학술대회논문집
    • /
    • 한국정보통신학회 2019년도 춘계학술대회
    • /
    • pp.313-315
    • /
    • 2019
  • 국내 패션 시장이 마이너스 성장을 전망하였으나 2018년도부터 다시 성장하고 있는 추세이다. 이러한 현상은 사람들이 패션에 대한 관심도가 증가하고 있다는 것을 의미한다. 패션에 대한 관심도가 증가함에 따라 사람들은 자신에게 어울리는 코디를 찾기 위해 여러 커뮤니티 사이트에 방문하여 참고하고 있다. 그러나 대부분의 커뮤니티 사이트들은 각각의 의류에 대해 카테고리 분류 작업을 수작업으로 하고 있다. 이러한 작업은 시간이 많이 소요될 뿐만이 아니라 여러 의류를 동시에 검색을 못 한다는 불편함이 있다. 즉 상의와 하의를 내가 원하는 것을 동시에 선택할 수 없고 상의를 선택하면 상의를 제외한 나머지 옷들은 모델이 입은 것을 보고 참고 해야 한다는 것이다. 이것에 대한 문제점은 제공되는 모델이 입은 코디가 사용자가 원하지 않는 코디일 확률이 높으므로 도움이 되지 않을 수 있다. 본 논문에서는 해당 문제점을 개선하기 위해 이미지를 업로드 하면 의류를 AI 분석 모델로 분석하여 자동으로 카테고리를 분류하여 저장한다. 따라서 기존의 방식대로 의류 1개로만 검색할 수 있을 뿐만이 아니라 자신이 원하는 의류를 여러 개 동시에 검색할 수 있다는 장점이 있다. 해당 서비스를 통해 기존보다 더 많은 사람들이 손쉽게 자신에게 맞는 코디를 찾아 참고할 수 있을 것으로 기대된다.

  • PDF

꼭두서니에 의한 면직물의 염색시 매염제와 키토산 처리가 색상에 미치는 영향 (Effect of Mordant Concentration and Chitosan Treatment on Dyeing Property)

  • 최정임;전동원
    • 한국의류산업학회지
    • /
    • 제5권3호
    • /
    • pp.283-288
    • /
    • 2003
  • Large quantity of metal mordant in natural dyeing has been used. For environmental reason, there is a need to reduce the amount of metal mordant and suggest the optimum amount of metal mordant in dyeing process. In this study, the relationship between mordant concentration and dyeing property was investigated. Various metal mordant concentrations of AI, Cu, Fe, Sn, were studied (from 1 to 5%). After treatment of mordant on cotton fabric, the concentration of metal ion in used mordant solution was quantified. There was no significant difference of the tone of color in mordant concentration from 1% to 5%, Therefore, there is no need to increase mordant concentration over 1%. chitosan treated cotton fabric showed a significant increase in dyeing affinity compared to untreated cotton fabric (the value of ${\Delta}E$ was 20). chitosan increased dye affinity significantly in the presence of mordant. It can be suggested that chitosan itself can replace metal mordants in the future.

이미지 기반 가상 착용 이미지 합성 기술 동향 (A Survey of Image-based Virtual Try-on Technology)

  • 박순찬;박진아;박지영
    • 전자통신동향분석
    • /
    • 제39권3호
    • /
    • pp.107-115
    • /
    • 2024
  • Image synthesis has been remarkably developed in the computer vision domain and various researches have been proposed to generate realistic and high-resolution images. In particular, image-based virtual try-on is an application in fashion domain to simulate wearing clothes. Specifically, using input images of a fashion model and products, an realistic image of the model wearing the provided garments is synthesized. In this paper, we present a comprehensive review of technical trends in image-based virtual try-on technology. We first introduce relevant datasets and discuss their characteristics. Then, we categorize existing image synthesis methods into three main streams: warping-based methods, encoding-decoding-based methods, and diffusion-based methods. Finally, we explore other important research issues in the field of virtual try-on and analyze related researches aimed to tackling those challenges.

AI Comparative Analysis of Trade and Consumption Patterns in Korea and China

  • Chang Hwan Choi;Thi Thanh Tuyen Nguyen;PengYan Wang
    • Journal of Korea Trade
    • /
    • 제27권1호
    • /
    • pp.119-138
    • /
    • 2023
  • Purpose - This research is to empirically explore the differences in apparel consumption among male and female teenagers and college students in Korea and China. By conducting a survey to understand customers' needs and behaviors, fashion businesses will be able to improve their customer satisfaction and avoid redundancy, inventory, and the waste of resources, effort and money. Design/methodology - The research design considers the consumption patterns of male and female high school and college students in Korea and China. To analyze the data, the study employs decision trees, a type of machine learning algorithm. A decision tree model was developed to examine the relationship between the explanatory and response variables, which can be either quantitative or qualitative in nature. Findings - The main findings of this study indicate that there are differences in shopping behavior among different customer segments. The results show that men have a simpler shopping behavior compared to women. Additionally, cultural factors and the difference in fashion needs between students and non-students have a significant impact on the shopping choices of Chinese and Korean individuals. Originality/value - Existing studies often assume that the shopping behavior of high school and university students is similar and that there are no significant differences in clothing purchases between men and women across countries. The results provide valuable insights into the unique shopping behavior of different customer segments, and can inform fashion businesses in their efforts to meet the needs of their customers.

메트로폴리탄 박물관의 18세기 복식전시가 현대 패션에 미친 영향 연구 (A Study on the Influence of 18th Century Costumes in Contemporary Fashion)

  • 윤은재;박형애
    • 대한가정학회지
    • /
    • 제44권1호
    • /
    • pp.25-35
    • /
    • 2006
  • This study investigated the scheme for correctly making Korean fashion design known to the world. It attempted to increase the influence of the Eighteenth Century Costume in contemporary fashion. During the 18th century, France had an almost complete monopoly of fashion. Growing out the fairyland atmosphere of the French Court and often conceived of as ennui by personal vanity, this fashion was a product of an age which sought at any price to live life with supreme grace. Most of the special costume exhibitions in the Metropolitan Museum of Art are planned and directed by Polaire Weissman, Diana Vreeland, Richard Martin, Harold Koda. The Costume Institute has held exhibitions of the Eighteenth Century Costume several times such as "Museum Period Rooms Re-Occupied in Style," "the Eighteenth Century Women," "the Ceaseless Century," "Dangerous Liaisons," etc. Especially, the exhibition of "Dangerous Liaisons" is organized in ten parts such as the Portrait, the Levee, the Music Lesson, the Withdrawing Room, the Broken Vase, the Favorite, the Masked Beauty, the Card Game, the Late Supper, and the Shop. Using the eighteenth century as its touchstone, The Ceaseless Century proceeds differently, not seeking the short distance between a discrete present and the multiple past but rather showing the complicated navigation that comes of revivalism swing to and fro on the timeline of history and sensibility. The designers featured include Karl Lagerfeld, Gianni Versace, Vivienne Westwood, Jean Paul Gaultier, Christian Dior, Cristobal Balencicga, Christian Lacroix, Stella McCartney forChloe, Olivier Theyskens, Alexander McQueen, etc. Therefore, Korean designers should refrain from (Ed-confirm) the foreign collection without a clear purpose and should devote their effort to create with an active attitude.

비정형 패션 이미지 검색을 위한 MASK R-CNN 선형처리 기반 CNN 분류 학습모델 구현 (Implementation of CNN-based Classification Training Model for Unstructured Fashion Image Retrieval using Preprocessing with MASK R-CNN)

  • 조승아;이하영;장혜림;김규리;이현지;손봉기;이재호
    • 한국산업정보학회논문지
    • /
    • 제27권6호
    • /
    • pp.13-23
    • /
    • 2022
  • 본 논문에서는 패션 분야의 비정형 데이터 검색을 위한 패션 아이템별 세부 컨포넌트 이미지 분류 알고리즘을 제안한다. 코로나-19 환경으로 인하여 최근 AI 기반 쇼핑몰이 증가하는 추세이다. 하지만 기존의 키워드 검색과 사용자 서핑 행위 기반 개인 맞춤형 스타일 추천으로는 정확한 비정형 데이터 검색에는 한계가 있다. 본 연구는 다양한 온라인 쇼핑 사이트에서 크롤링한 이미지를 사용하여 Mask R-CNN을 활용한 전처리를 진행한 후, CNN을 통해 패션 아이템별 컴포넌트에 대한 분류를 진행하였다. 셔츠의 카라 및 패턴과 청바지의 핏, 워싱 및 컬러에 대한 분류를 진행하였으며, 다양한 전이학습 모델을 비교 분석한 후 가장 높은 정확도가 나온 Densenet121모델을 사용하여 셔츠의 카라는 93.28%, 셔츠의 패턴은 98.10%의 정확도를 도달하였으며, 청바지의 핏은 Notched, Spread, Straight 3가지의 클래스의 경우 91.73%, Regular 핏을 추가한 4가지의 클래스의 경우 81.59%, 청바지의 색상은 93.91%, 청바지의 Washing은 91.20%, 청바지의 Demgae는 92.96%의 정확도를 도출하였다.

통영오광대 복식의 의도성에 관한 연구 : 대사를 중심으로 (A Study on Intentions of Tongyeng Ogwangdae Costume : Focused on dialogs)

  • 김초영;김은정
    • 한국의류산업학회지
    • /
    • 제17권4호
    • /
    • pp.541-549
    • /
    • 2015
  • Korean traditional mask play was done at outdoor to let spectators join the play and take mask at performance and to have disadvantage of poor transfer of dialog to the spectators. Performers transferred dialog directly and/or indirectly by using visual costume, and Tongyeong Ogwangdae with many dialogue expressions concerning the costume made use of proper intention to help spectators understand. In this study, the author watched performance of Tongyeong Ogwangdae to collect and investigate not only image material but also photograph material and to classify intention of the costume into three: First, contents of the dialog were intended to make shape. The costumes of Hongbaek Yangban, Meoktal Yangban and Jorijung of the 2nd section gave spectators contents of the play, and the costume of Halmi Yangban of Nochangtal of the 4th section described dialog of Halmi. Second, dialog had connotative metaphor. The costume of Yeongno of Yeongnotal of the 3rd section symbolized an unfinished dragon, while that of Halmi of Nongchangtal of the 4th section did sacred of Halmi. And, Nongchagtal Ai's costume of the 4th section prayed for value of Ai, richness of the life and fecundity. Third, conflict between characters was intended by indirect media. The costume of Maltukee of metaphor mask of the 2nd section became base of the assertion supporting that Maltukee had better birth history than Yangban had, and the costume of Yeongnoyangban of Yeongnotal of the 3rd section produced conflict at disclosure of identity.