메뉴 건너뛰기
Library Notice
Institutional Access
If you certify, you can access the articles for free.
Check out your institutions.
ex)Hankuk University, Nuri Motors
Log in Register Help KOR
Subject

Subtitle Automatic Extraction System for Short-form Contents
Recommendations
Search
Questions

숏폼 콘텐츠 자막 자동 추출 시스템

논문 기본 정보

Type
Academic journal
Author
Junyoung Jo (군산대학교) Jangwon Gim (군산대학교) Byung-Won On (군산대학교) Dongwon Jeong (군산대학교)
Journal
Korean Institute of Information Technology The Journal of Korean Institute of Information Technology Vol.19 No.6 KCI Accredited Journals
Published
2021.6
Pages
29 - 37 (9page)
DOI
10.14801/jkiit.2021.19.6.29

Usage

cover
📌
Topic
📖
Background
🔬
Method
🏆
Result
Subtitle Automatic Extraction System for Short-form Contents
Ask AI
Recommendations
Search
Questions

Abstract· Keywords

Report Errors
Recently, short-form content, which means short and intense content, is rapidly increasing in YouTube and mobile platform environments. Short-form contents use special subtitles and general subtitles composed of a combination of images (objects, people) and Hangul letters to deliver key content to viewers in a short time and give high satisfaction compared to time investment. Therefore, a particular subtitle extraction method is needed in short-form contents. In this paper, we propose an automatic identification system for subtitles appearing in short-form content. Through the algorithm for identifying the shape of the subtitles of the proposed system, the subtitle extraction accuracy (95%) included in the short-form contents was shown to be excellent. Therefore, it is expected that the proposed system can apply to the user recommendation system and content curation service for short-form content in the future.

Contents

요약
Abstract
Ⅰ. 서론
Ⅱ. 관련 동향
Ⅲ. 제안 시스템
Ⅳ. 실험 및 평가
Ⅴ. 결론 및 향후 과제
References

References (16)

Add References

Recommendations

It is an article recommended by DBpia according to the article similarity. Check out the related articles!

Related Authors

Frequently Viewed Together

Recently viewed articles

Comments(0)

0

Write first comments.