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
학술대회자료
저자정보
정민영 (연세대학교) 이현수 (연세대학교)
저널정보
한국실내디자인학회 한국실내디자인학회 학술대회논문집 한국실내디자인학회 2022년도 춘계학술발표대회 논문집
발행연도
2022.5
수록면
111 - 116 (6page)

이용수

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

이 논문의 연구 히스토리 (2)

초록· 키워드

오류제보하기
This research suggests recommending modern artwork using user-centered hash tagged Instagram images. The aim of this study was twofold. First, it is an attempt to overcome the limitation of over fitting data with the daily image on Instagram. Therefore, this research verifies daily images as one classified style. Second, it is an approach for predicting quantified image data with users’ Hashtag. user-centered hashtag is a way to curate artwork for a user. This research is based on the Deep learning method for hashtag quantification that can replace thousands of people’s daily styles and cognition. In the proposed simulation, Scraping Images with top-rated hashtag as quantified data. Hash tags - Cute, Fashion, Natural, Food - are firstly processed in the frequency domain to artwork images which can be treated as a hashtag classification. Selfie and advertisements are distracted as pre-processing images. For test data, a thousand painting artwork posts of MoMA (Museum of Modern Art) were collected too. Search pre-processed hashtag daily images to get artwork images fed in CNN (Convolutional Neural Network) architecture. For artwork image clustered, hashtag daily images are fed in CNN. In the final stage of Transfer Learning, Images are transfered bottlenecks classified from each Hashtag image folder. The output of the CNNs are fused using TF(TensorFlow) API. The input of the fusion is given to support the Python language for image classification. The proposed system is evaluated using the Tensor board, the proven data. Ten thousand lists have been classified. For usability, Top-rated images are automatically extracted on an excel file coded for this research with Python language. This research has concluded that it is desirable to use TF for predicting the set of Images from hashtag and it helps for recommendations efficiently. TF works for the hash tagged image classifying artwork images. Modern artwork image is classified using deep learning, and analysis of hashtag. Cute, Fashion, Natural - classified each hash-tagged artwork in a minute for a thousand Instagram images of MoMA more than 90 percent. There was a limitation to the hashtag. there is no painting style for Food hashtag with more than 60 percent. Therefore, it is expected to use determine which style of art can be found too without over fitting data using hash tag images.

목차

Abstract
1. 서론
2. 이론적 배경
3. 결과분석
4. 결론
참고문헌

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

UCI(KEPA) : I410-ECN-0101-2023-619-000155345