解码漫画视觉叙事：使用信息建模技术的半自动回顾

IF 2.4 3区计算机科学 Q2 COMPUTER SCIENCE, CYBERNETICS

Entertainment Computing Pub Date : 2025-07-09 DOI:10.1016/j.entcom.2025.100995

Rishu, Vinay Kukreja

{"title":"解码漫画视觉叙事：使用信息建模技术的半自动回顾","authors":"Rishu, Vinay Kukreja","doi":"10.1016/j.entcom.2025.100995","DOIUrl":null,"url":null,"abstract":"<div><h3>Context</h3><div>Comic analysis has become an increasingly prominent area because of the rise in digital comics. It focuses on automatically identifying and extracting comic panels, speech balloons, and text from comics, with various applications in the digital media sector.</div></div><div><h3>Objective</h3><div>The study investigated the research trends and semantic relationship between qualitative and quantitative literature on comic analysis.</div></div><div><h3>Method</h3><div>A semi-automatic literature review was conducted on 493 research studies (2005–2024), retrieved from major academic databases. Latent semantic analysis (LSA) method was selected over other information modeling techniques due to its ability to handle synonymy, polysemy, and sparse datasets.</div></div><div><h3>Result</h3><div>The study identifies leading researchers such as Jean-Christophe Burie, Christophe Rigaud, Yusuke Matsui, and Clement Guerin (28.63%), and the top publishers include Journal of Graphic Novels and Comics, Studies in Comics, and International Conference on Document Analysis and Recognition (25.60%). Speech balloon analysis (21.4%), panel detection (18.9%), visual narrative patterns (16.7%), automatic comic text extraction (12.3%), and semantic storytelling analysis (10.6%) are the most researched topics. A five-topic solution (coherence score: 0.847) was generated, and validated via Cohen’s Kappa (0.950) for inter-rater reliability, ensuring robust topic assignments. These insights benefit AI-driven recommendation systems, digital media platforms, and multimodal content analysis.</div></div>","PeriodicalId":55997,"journal":{"name":"Entertainment Computing","volume":"55 ","pages":"Article 100995"},"PeriodicalIF":2.4000,"publicationDate":"2025-07-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":"{\"title\":\"Decoding comics visual narratives: a semi-automatic review using information modelling technique\",\"authors\":\"Rishu, Vinay Kukreja\",\"doi\":\"10.1016/j.entcom.2025.100995\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"<div><h3>Context</h3><div>Comic analysis has become an increasingly prominent area because of the rise in digital comics. It focuses on automatically identifying and extracting comic panels, speech balloons, and text from comics, with various applications in the digital media sector.</div></div><div><h3>Objective</h3><div>The study investigated the research trends and semantic relationship between qualitative and quantitative literature on comic analysis.</div></div><div><h3>Method</h3><div>A semi-automatic literature review was conducted on 493 research studies (2005–2024), retrieved from major academic databases. Latent semantic analysis (LSA) method was selected over other information modeling techniques due to its ability to handle synonymy, polysemy, and sparse datasets.</div></div><div><h3>Result</h3><div>The study identifies leading researchers such as Jean-Christophe Burie, Christophe Rigaud, Yusuke Matsui, and Clement Guerin (28.63%), and the top publishers include Journal of Graphic Novels and Comics, Studies in Comics, and International Conference on Document Analysis and Recognition (25.60%). Speech balloon analysis (21.4%), panel detection (18.9%), visual narrative patterns (16.7%), automatic comic text extraction (12.3%), and semantic storytelling analysis (10.6%) are the most researched topics. A five-topic solution (coherence score: 0.847) was generated, and validated via Cohen’s Kappa (0.950) for inter-rater reliability, ensuring robust topic assignments. These insights benefit AI-driven recommendation systems, digital media platforms, and multimodal content analysis.</div></div>\",\"PeriodicalId\":55997,\"journal\":{\"name\":\"Entertainment Computing\",\"volume\":\"55 \",\"pages\":\"Article 100995\"},\"PeriodicalIF\":2.4000,\"publicationDate\":\"2025-07-09\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"0\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"Entertainment Computing\",\"FirstCategoryId\":\"94\",\"ListUrlMain\":\"https://www.sciencedirect.com/science/article/pii/S1875952125000758\",\"RegionNum\":3,\"RegionCategory\":\"计算机科学\",\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"Q2\",\"JCRName\":\"COMPUTER SCIENCE, CYBERNETICS\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"Entertainment Computing","FirstCategoryId":"94","ListUrlMain":"https://www.sciencedirect.com/science/article/pii/S1875952125000758","RegionNum":3,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q2","JCRName":"COMPUTER SCIENCE, CYBERNETICS","Score":null,"Total":0}

引用次数: 0

摘要

由于数字漫画的兴起，漫画分析已经成为一个日益突出的领域。它专注于自动识别和提取漫画面板、演讲气球和漫画文本，在数字媒体领域有各种应用。目的探讨漫画分析定性文献和定量文献的研究趋势和语义关系。方法对2005-2024年间从各大学术数据库中检索的493篇研究进行半自动文献综述。由于潜在语义分析（LSA）方法具有处理同义、多义和稀疏数据集的能力，因此在其他信息建模技术中被选择。结果本研究确定了Jean-Christophe Burie、Christophe Rigaud、Yusuke Matsui和Clement Guerin等主要研究人员（28.63%），以及Journal of Graphic novel and Comics、Studies in Comics和International Conference on Document Analysis and Recognition（25.60%）等顶级出版商。语音气球分析（21.4%）、面板检测（18.9%）、视觉叙事模式（16.7%）、自动漫画文本提取（12.3%）和语义故事分析（10.6%）是研究最多的主题。生成了一个五主题解决方案（一致性得分：0.847），并通过Cohen 's Kappa（0.950）验证了评分者之间的信度，确保了稳健的主题分配。这些见解有利于人工智能驱动的推荐系统、数字媒体平台和多模式内容分析。

本文章由计算机程序翻译，如有差异，请以英文原文为准。

查看原文本刊更多论文

Decoding comics visual narratives: a semi-automatic review using information modelling technique

Context

Comic analysis has become an increasingly prominent area because of the rise in digital comics. It focuses on automatically identifying and extracting comic panels, speech balloons, and text from comics, with various applications in the digital media sector.

Objective

The study investigated the research trends and semantic relationship between qualitative and quantitative literature on comic analysis.

Method

A semi-automatic literature review was conducted on 493 research studies (2005–2024), retrieved from major academic databases. Latent semantic analysis (LSA) method was selected over other information modeling techniques due to its ability to handle synonymy, polysemy, and sparse datasets.

Result

The study identifies leading researchers such as Jean-Christophe Burie, Christophe Rigaud, Yusuke Matsui, and Clement Guerin (28.63%), and the top publishers include Journal of Graphic Novels and Comics, Studies in Comics, and International Conference on Document Analysis and Recognition (25.60%). Speech balloon analysis (21.4%), panel detection (18.9%), visual narrative patterns (16.7%), automatic comic text extraction (12.3%), and semantic storytelling analysis (10.6%) are the most researched topics. A five-topic solution (coherence score: 0.847) was generated, and validated via Cohen’s Kappa (0.950) for inter-rater reliability, ensuring robust topic assignments. These insights benefit AI-driven recommendation systems, digital media platforms, and multimodal content analysis.

求助全文

通过发布文献求助，成功后即可免费获取论文全文。去求助

来源期刊

Entertainment Computing Computer Science-Human-Computer Interaction

CiteScore

5.90

自引率

7.10%

发文量

期刊介绍： Entertainment Computing publishes original, peer-reviewed research articles and serves as a forum for stimulating and disseminating innovative research ideas, emerging technologies, empirical investigations, state-of-the-art methods and tools in all aspects of digital entertainment, new media, entertainment computing, gaming, robotics, toys and applications among researchers, engineers, social scientists, artists and practitioners. Theoretical, technical, empirical, survey articles and case studies are all appropriate to the journal.