[论文解读] InkIdeator: Supporting Chinese-Style Visual Design Ideation via AI-Infused Exploration of Chinese Paintings

Shiwei Wu, Ziyao Gao+7 more|arXiv (Cornell University)|Jan 26, 2026

Aesthetic Perception and Analysis被引用 0

一句话总结

InkIdeator 是一款AI辅助创意产生工具，通过对中国画进行文化符号、情感、构图与风格的注释来指导符号、情感、构图分析与视觉草图生成，已通过形成性研究与同质性研究进行验证。

ABSTRACT

Visual designers often seek inspiration from Chinese paintings when tasked with creating Chinese-style illustrations, posters, etc. Our formative study (N=10) reveals that during ideation, designers learn the cultural symbols, emotions, compositions, and styles in Chinese paintings but face challenges in searching, analyzing, and integrating these dimensions. This paper leverages multi-modal large models to annotate the value of each dimension in 16,315 Chinese paintings, built on which we propose InkIdeator, an ideation support system for Chinese-style visual designs. InkIdeator suggests cultural symbols associated with the task theme, provides dimensional keywords to help analyze Chinese paintings, and generates visual examples integrating user-selected keywords. Our within-subjects study (N=12) using a baseline system without extracted dimensional keywords, along with two extended use cases by Chinese painters, indicates InkIdeator's effectiveness in creative ideation support, helping users efficiently explore cultural dimensions in Chinese paintings and visualize their ideas. We discuss implications for supporting culture-related visual design ideation with generative AI.

研究动机与目标

• identify design challenges designers face when ideating with Chinese paintings.
• Create a richly annotated Chinese painting dataset capturing cultural symbols, emotions, compositions, and styles.
• Develop InkIdeator to help search, analyze, and visualize design elements from paintings for ideation.

提出的方法

• Conduct a formative study with 10 designers to identify ideation challenges and preferred dimensions in Chinese paintings.
• Build a dataset of 16,315 Chinese paintings annotated with cultural symbols, emotions, dimensions, and styles using a multimodal LLM (GPT-4o mini) prompted with Role Play, Dimension Analysis, Knowledge Injection and JSON outputs.
• Annotate and cluster cultural symbols and emotions; classify painting types (Gongbi/Xieyi) with a classifier trained on 500 images.
• Develop four interactive components (Symbol Association Panel, Image Library, Moodboard Canvas, Image Generation Panel) linked to the annotated dataset to support three iterative tasks: Search Reference, Analyze Reference, Visualize Ideas.
• Evaluate InkIdeator with a within-subjects study (N=12) against a baseline (ChatGPT + image search + MidJourney) and two case studies with Chinese painters.

实验结果

研究问题

• RQ1 AI 如何对中国画中的文化符号、情感、构图与风格进行注释和组织，以支持设计创意产生？
• RQ2 与基线相比，InkIdeator 是否提升设计师在检索、分析和可视化中国风视觉设计参考方面的能力？
• RQ3 经验丰富的中国画家是否能够有效使用 InkIdeator 完成超出设计师的创意任务？
• RQ4 当将生成式AI 融入文化特定的创意过程时，会出现哪些设计空间和工作流程方面的考虑？

主要发现

• InkIdeator 实现了对中国画视觉参考的有序探索以及从中提取设计元素的更高效性.
• 参与者认为 InkIdeator 在创意任务中提供的支持感超过基线。
• 该工具比基线更高效地将想法转化为视觉呈现。
• 两位经验丰富的中国画家在 InkIdeator 上对创意进行了迭代，表明该工具对与文化相关的创意任务具有潜在的泛化性。
• 基于 16,315 幅画作的带注释设计空间涵盖文化符号、情感、构图与风格，可用于引导检索与分析。