A USER-GUIDED GENERATIVE IMAGE-BASED MODEL INTERFACE TO DESIGN DESIRED PRODUCTS WITH ITERATIVE DESIGN SELECTION
Chair design as a case study
2025 年 9 巻 1 号 p. 1_11-1_20
詳細
2025 年 9 巻 1 号 p. 1_11-1_20
This study utilized high fidelity prototype to assess the user experience of interacting with a new stable diffusion image generator interface while users tried to generate the desired chair design through iterative selections. User Experience Questionnaire (UEQ) results of the experiment were compared with the UEQ benchmark containing a data set of 85 similar products to explore its user experience (UX) sufficiency in fulfilling respondents’ expectations. Dialogue failure recommendations by experts, based on the Evaluative Classification of Mismatch (ECM) method, elucidated respondent’s system requirements. The impact levels of dialogue failures judged by UX experts and analyzed with Fleiss’ Kappa analysis, results compared to UEQ results, and their relation estimation with prototype features were prioritized. Eventually, prominent interface principles applying to the generative image base interface in this research were determined. The current study helps image generative system designers in terms of how users interact with the system through the interface and their requirements.
