DesignWeaver: Dimensional Scaffolding for Text-to-Image Product Design

arXiv cs.AI / 3/30/2026

💬 OpinionSignals & Early TrendsIdeas & Deep AnalysisTools & Practical UsageModels & Research

共有:

Key Points

The paper studies how expert product designers explore and communicate design spaces, finding that they often rely on visual references and shared dimensions rather than purely written descriptions.
It introduces DesignWeaver, an interface that extracts key product-design dimensions from generated images and presents them as selectable options to help novices form better prompts for text-to-image models.
In a novice study with 52 participants, DesignWeaver helped users write longer prompts with more domain-specific vocabulary, leading to more diverse and innovative product design outputs.
The approach also caused users to develop higher expectations than current text-to-image models can reliably satisfy, highlighting a practical UX/expectation-management challenge for AI-assisted design tools.

Abstract

Generative AI has enabled novice designers to quickly create professional-looking visual representations for product concepts. However, novices have limited domain knowledge that could constrain their ability to write prompts that effectively explore a product design space. To understand how experts explore and communicate about design spaces, we conducted a formative study with 12 experienced product designers and found that experts -- and their less-versed clients -- often use visual references to guide co-design discussions rather than written descriptions. These insights inspired DesignWeaver, an interface that helps novices generate prompts for a text-to-image model by surfacing key product design dimensions from generated images into a palette for quick selection. In a study with 52 novices, DesignWeaver enabled participants to craft longer prompts with more domain-specific vocabularies, resulting in more diverse, innovative product designs. However, the nuanced prompts heightened participants' expectations beyond what current text-to-image models could deliver. We discuss implications for AI-based product design support tools.