InkSpirit：一种专业知识驱动的方法，用于增强传统中国画文本到图像生成的视觉逻辑

IF 2.8 4区计算机科学 Q2 COMPUTER SCIENCE, SOFTWARE ENGINEERING

Computers & Graphics-Uk Pub Date : 2025-08-13 DOI:10.1016/j.cag.2025.104330

Di Zhang , Chen Yi , Xinyu Gao , Xiangsheng Zeng , Runqiao Xia , Yongbo Jiang , Yingchaojie Feng , Wei Zhang , Wei Chen

{"title":"InkSpirit：一种专业知识驱动的方法，用于增强传统中国画文本到图像生成的视觉逻辑","authors":"Di Zhang , Chen Yi , Xinyu Gao , Xiangsheng Zeng , Runqiao Xia , Yongbo Jiang , Yingchaojie Feng , Wei Zhang , Wei Chen","doi":"10.1016/j.cag.2025.104330","DOIUrl":null,"url":null,"abstract":"<div><div>Traditional Chinese painting (TCP) presents unique challenges for text-to-image models, including composition logic deficiency, lack of inscription semantics, and style deviations. This study proposes the “InkSpirit” framework, employing an expert knowledge-driven approach to address these issues by: (1) constructing a TCP dataset with composition-based Blank Space Principles, (2) building an Artistic conception-Inscription corpus, and (3) designing a generation framework based on ComfyUI workflow for precise control over TCP elements. Experiments demonstrate superior performance in image quality metrics, with validation through expert and user evaluations, advancing the integration of traditional art with AI technology.</div></div>","PeriodicalId":50628,"journal":{"name":"Computers & Graphics-Uk","volume":"132 ","pages":"Article 104330"},"PeriodicalIF":2.8000,"publicationDate":"2025-08-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":"{\"title\":\"InkSpirit: An expert knowledge-driven approach for enhancing the visual logic of traditional Chinese painting text-to-image generation\",\"authors\":\"Di Zhang , Chen Yi , Xinyu Gao , Xiangsheng Zeng , Runqiao Xia , Yongbo Jiang , Yingchaojie Feng , Wei Zhang , Wei Chen\",\"doi\":\"10.1016/j.cag.2025.104330\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"<div><div>Traditional Chinese painting (TCP) presents unique challenges for text-to-image models, including composition logic deficiency, lack of inscription semantics, and style deviations. This study proposes the “InkSpirit” framework, employing an expert knowledge-driven approach to address these issues by: (1) constructing a TCP dataset with composition-based Blank Space Principles, (2) building an Artistic conception-Inscription corpus, and (3) designing a generation framework based on ComfyUI workflow for precise control over TCP elements. Experiments demonstrate superior performance in image quality metrics, with validation through expert and user evaluations, advancing the integration of traditional art with AI technology.</div></div>\",\"PeriodicalId\":50628,\"journal\":{\"name\":\"Computers & Graphics-Uk\",\"volume\":\"132 \",\"pages\":\"Article 104330\"},\"PeriodicalIF\":2.8000,\"publicationDate\":\"2025-08-13\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"0\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"Computers & Graphics-Uk\",\"FirstCategoryId\":\"94\",\"ListUrlMain\":\"https://www.sciencedirect.com/science/article/pii/S0097849325001724\",\"RegionNum\":4,\"RegionCategory\":\"计算机科学\",\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"Q2\",\"JCRName\":\"COMPUTER SCIENCE, SOFTWARE ENGINEERING\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"Computers & Graphics-Uk","FirstCategoryId":"94","ListUrlMain":"https://www.sciencedirect.com/science/article/pii/S0097849325001724","RegionNum":4,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q2","JCRName":"COMPUTER SCIENCE, SOFTWARE ENGINEERING","Score":null,"Total":0}

引用次数: 0

摘要

传统国画在文象模型上面临着独特的挑战，包括构图逻辑缺陷、题词语义缺失和风格偏差。本研究提出了“InkSpirit”框架，采用专家知识驱动的方法来解决这些问题：(1)构建基于构图的空白原则的TCP数据集，(2)构建一个艺术概念-铭文语料库，(3)设计一个基于ComfyUI工作流的生成框架，以精确控制TCP元素。实验表明，在图像质量指标方面表现优异，并通过专家和用户评估进行验证，推进了传统艺术与人工智能技术的融合。

本文章由计算机程序翻译，如有差异，请以英文原文为准。

InkSpirit: An expert knowledge-driven approach for enhancing the visual logic of traditional Chinese painting text-to-image generation

查看原文本刊更多论文

InkSpirit: An expert knowledge-driven approach for enhancing the visual logic of traditional Chinese painting text-to-image generation

Traditional Chinese painting (TCP) presents unique challenges for text-to-image models, including composition logic deficiency, lack of inscription semantics, and style deviations. This study proposes the “InkSpirit” framework, employing an expert knowledge-driven approach to address these issues by: (1) constructing a TCP dataset with composition-based Blank Space Principles, (2) building an Artistic conception-Inscription corpus, and (3) designing a generation framework based on ComfyUI workflow for precise control over TCP elements. Experiments demonstrate superior performance in image quality metrics, with validation through expert and user evaluations, advancing the integration of traditional art with AI technology.

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

Computers & Graphics-Uk 工程技术-计算机：软件工程

CiteScore

5.30

自引率

12.00%

发文量

173

审稿时长

38 days

期刊介绍： Computers & Graphics is dedicated to disseminate information on research and applications of computer graphics (CG) techniques. The journal encourages articles on: 1. Research and applications of interactive computer graphics. We are particularly interested in novel interaction techniques and applications of CG to problem domains. 2. State-of-the-art papers on late-breaking, cutting-edge research on CG. 3. Information on innovative uses of graphics principles and technologies. 4. Tutorial papers on both teaching CG principles and innovative uses of CG in education.