메뉴 건너뛰기
.. 내서재 .. 알림
소속 기관/학교 인증
인증하면 논문, 학술자료 등을  무료로 열람할 수 있어요.
한국대학교, 누리자동차, 시립도서관 등 나의 기관을 확인해보세요
(국내 대학 90% 이상 구독 중)
로그인 회원가입 고객센터 ENG
주제분류

추천
검색
질문

논문 기본 정보

자료유형
학술저널
저자정보
Tatsuki Oike (Sophia University) Haruka Yamashita (Sophia University) Ryotaro Shimizu (Waseda University)
저널정보
대한산업공학회 Industrial Engineering & Management Systems Industrial Engineering & Management Systems Vol.24 No.1
발행연도
2025.3
수록면
21 - 28 (8page)
DOI
10.7232/iems.2025.24.1.021

이용수

표지
📌
연구주제
📖
연구배경
🔬
연구방법
🏆
연구결과
AI에게 요청하기
추천
검색
질문

초록· 키워드

오류제보하기
In recent years, image-generation technology has been successfully applied in the fashion industry. Previous studies have focused on the generation of fashion images that specify the clothes and poses of the person in the image or the changing of the clothes and poses of existing images. However, images were not generated with consideration given to attributes such as gender or height. To adapt image generation to real-world services, it is essential to generate images based on such user attributes. Consequently, in this study we generated images by utilizing the data of an actual fashion coordination site—that is, WEAR operated by ZOZO, Inc—with consideration given to attributes such as the gender, and height of the person in the image and an evaluation index of the image called the “number of likes”. First, a class label for each image was created based on the attribute information of the person in the image and evaluation of the image. By applying the created class label to the conditional part of the conditional StyleGAN2-ADA model and learning its parameters, it was possible to generate a “highly rated image” that was likely to be evaluated many times considering the attributes of the person in the image. Furthermore, visual inspection was conducted to evaluate whether an image with the specified attribute was actually generated, and image classification using EfficientNet V2 was performed to confirm whether a “highly evaluated image” had been generated.

목차

ABSTRACT
1. INTRODUCTION
2. RELATED WORK
3. CREATION OF CLASS LABELS
4. EXPERIMENTS
5. DISCUSSIONS
6. CONCLUSION AND FUTURE WORKS
REFERENCES

참고문헌 (0)

참고문헌 신청

함께 읽어보면 좋을 논문

논문 유사도에 따라 DBpia 가 추천하는 논문입니다. 함께 보면 좋을 연관 논문을 확인해보세요!

이 논문의 저자 정보

최근 본 자료

전체보기

댓글(0)

0