Conceptual framework of hybrid style in fashion image datasets for machine learning
Abstract Fashion image datasets, in which each fashion image has a label indicating its design attributes and styles, have contributed to the achievement of various machine learning techniques in the fashion industry. Computer vision studies have investigated labeling categories (such as fashion ite...
Main Authors: | , , , |
---|---|
Format: | Article |
Language: | English |
Published: |
SpringerOpen
2023-05-01
|
Series: | Fashion and Textiles |
Subjects: | |
Online Access: | https://doi.org/10.1186/s40691-023-00338-8 |
_version_ | 1797823018428268544 |
---|---|
author | Hyosun An Kyo Young Lee Yerim Choi Minjung Park |
author_facet | Hyosun An Kyo Young Lee Yerim Choi Minjung Park |
author_sort | Hyosun An |
collection | DOAJ |
description | Abstract Fashion image datasets, in which each fashion image has a label indicating its design attributes and styles, have contributed to the achievement of various machine learning techniques in the fashion industry. Computer vision studies have investigated labeling categories (such as fashion items, colors, materials, details, and styles) to create fashion image datasets for supervised learning. Although a considerable number of fashion image datasets has been developed, different style classification criteria exist because of a lack of understanding concerning fashion style. Since fashion styles reflect various design attributes, multiple styles can often be included in a single outfit. Thus, this study aims to build a Hybrid Style Framework to develop a fashion image dataset that can be efficiently applied to supervised learning. We conducted focus group interviews with six fashion experts to determine fashion style categories with which to classify hybrid styles in fashion images. We developed 1,206,931K-fashion image datasets and analyzed the hybrid style convergence. Finally, we applied the datasets to the machine learning model and verified the accuracy of the computer’s ability to recognize style. Overall, this study concludes that the Hybrid Style Framework and developed K-fashion image datasets are helpful, as they can be applied to data-driven fashion services to offer personalized fashion design solutions. |
first_indexed | 2024-03-13T10:17:30Z |
format | Article |
id | doaj.art-725d5a66abf5496abc4f75387e82405b |
institution | Directory Open Access Journal |
issn | 2198-0802 |
language | English |
last_indexed | 2024-03-13T10:17:30Z |
publishDate | 2023-05-01 |
publisher | SpringerOpen |
record_format | Article |
series | Fashion and Textiles |
spelling | doaj.art-725d5a66abf5496abc4f75387e82405b2023-05-21T11:09:10ZengSpringerOpenFashion and Textiles2198-08022023-05-0110111810.1186/s40691-023-00338-8Conceptual framework of hybrid style in fashion image datasets for machine learningHyosun An0Kyo Young Lee1Yerim Choi2Minjung Park3Department of Fashion Industry, Ewha Womans UniversityDepartment of Fashion Industry, Ewha Womans UniversityDepartment of Data Science, Seoul Women’s UniversityDepartment of Fashion Industry, Ewha Womans UniversityAbstract Fashion image datasets, in which each fashion image has a label indicating its design attributes and styles, have contributed to the achievement of various machine learning techniques in the fashion industry. Computer vision studies have investigated labeling categories (such as fashion items, colors, materials, details, and styles) to create fashion image datasets for supervised learning. Although a considerable number of fashion image datasets has been developed, different style classification criteria exist because of a lack of understanding concerning fashion style. Since fashion styles reflect various design attributes, multiple styles can often be included in a single outfit. Thus, this study aims to build a Hybrid Style Framework to develop a fashion image dataset that can be efficiently applied to supervised learning. We conducted focus group interviews with six fashion experts to determine fashion style categories with which to classify hybrid styles in fashion images. We developed 1,206,931K-fashion image datasets and analyzed the hybrid style convergence. Finally, we applied the datasets to the machine learning model and verified the accuracy of the computer’s ability to recognize style. Overall, this study concludes that the Hybrid Style Framework and developed K-fashion image datasets are helpful, as they can be applied to data-driven fashion services to offer personalized fashion design solutions.https://doi.org/10.1186/s40691-023-00338-8Fashion imageHybrid styleMachine learningK-fashionSupervised learning datasets |
spellingShingle | Hyosun An Kyo Young Lee Yerim Choi Minjung Park Conceptual framework of hybrid style in fashion image datasets for machine learning Fashion and Textiles Fashion image Hybrid style Machine learning K-fashion Supervised learning datasets |
title | Conceptual framework of hybrid style in fashion image datasets for machine learning |
title_full | Conceptual framework of hybrid style in fashion image datasets for machine learning |
title_fullStr | Conceptual framework of hybrid style in fashion image datasets for machine learning |
title_full_unstemmed | Conceptual framework of hybrid style in fashion image datasets for machine learning |
title_short | Conceptual framework of hybrid style in fashion image datasets for machine learning |
title_sort | conceptual framework of hybrid style in fashion image datasets for machine learning |
topic | Fashion image Hybrid style Machine learning K-fashion Supervised learning datasets |
url | https://doi.org/10.1186/s40691-023-00338-8 |
work_keys_str_mv | AT hyosunan conceptualframeworkofhybridstyleinfashionimagedatasetsformachinelearning AT kyoyounglee conceptualframeworkofhybridstyleinfashionimagedatasetsformachinelearning AT yerimchoi conceptualframeworkofhybridstyleinfashionimagedatasetsformachinelearning AT minjungpark conceptualframeworkofhybridstyleinfashionimagedatasetsformachinelearning |