CreativeConnect: Supporting Reference Recombination for Graphic Design Ideation with Generative AI

DaEun Choi,Sumin Hong,Jeongeon Park,John Joon Young Chung,Juho Kim

2024-03-07

Abstract:Graphic designers often get inspiration through the recombination of references. Our formative study (N=6) reveals that graphic designers focus on conceptual keywords during this process, and want support for discovering the keywords, expanding them, and exploring diverse recombination options of them, while still having room for designers' creativity. We propose CreativeConnect, a system with generative AI pipelines that helps users discover useful elements from the reference image using keywords, recommends relevant keywords, generates diverse recombination options with user-selected keywords, and shows recombinations as sketches with text descriptions. Our user study (N=16) showed that CreativeConnect helped users discover keywords from the reference and generate multiple ideas based on them, ultimately helping users produce more design ideas with higher self-reported creativity compared to the baseline system without generative pipelines. While CreativeConnect was shown effective in ideation, we discussed how CreativeConnect can be extended to support other types of tasks in creativity support.

Human-Computer Interaction

What problem does this paper attempt to address?

The paper aims to address the issue of generating creative inspiration in the field of graphic design, particularly the process of stimulating new design ideas through the recombination of reference images. Specifically, the paper focuses on the challenges faced by designers during the conceptual creative stage and proposes a generative artificial intelligence (AI)-based approach to support this process.

The paper points out that in the creative process of graphic design, designers usually need to extract elements or aspects from multiple reference images, then consider the connections between these elements, and further integrate these elements to generate new design ideas. However, each step in this process requires designers to invest a significant amount of effort, including decomposing reference images, analyzing elements, exploring different combinations of elements, and so on.

To address the above challenges, the researchers developed a system called CreativeConnect. This system uses generative AI technology to help users:
1. **Easily extract key elements from reference images**: including main objects, actions and poses, themes and atmospheres, and composition aspects.
2. **Recommend related elements**: helping users discover more keywords related to the extracted elements to expand their creative space.
3. **Generate multiple recombination options**: generating diverse recombination schemes based on the keywords selected by the user and presenting them in the form of sketches and descriptions.
4. **Maintain a partially unfinished state**: to stimulate users' creativity, the generated results intentionally retain a certain degree of incompleteness.

Through empirical research, the authors demonstrated that CreativeConnect can effectively support designers in discovering elements and generating design ideas during the reference recombination process, helping designers generate more design ideas within a given time frame. Additionally, participants felt that their design ideas were more creative after using CreativeConnect. Furthermore, the study emphasized that low-fidelity sketch outputs help stimulate users' imagination, thereby further promoting the generation of creativity.