Design Exploration System based on Interior Design Network

Abstract

Designers develop concepts in the early design process to determine the most important aspects of the design. Designers search design references for design solutions during concept development. Sometimes the process of navigating through a large number of references requires a lot of effort on the part of the designer. In this context, design computing should provide design feedback that can effectively support the design process. Design computing should also focus on supporting the designer's behavior through interaction with the designer. Therefore, this thesis proposes a system that can support the designer's efficient design process by providing effective feedback suit to the designer's behavior in the design situation. To this end, this thesis conducted two major studies. First, I created an interior design dataset that can be retrieved to provide designers with appropriate feedback. This is a value derived by conducting a detailed analysis of interior design by design element. Next, I propose a BIG reference system that can provide various types of feedback through the dataset. The system provides similar feedback as the existing search system but provides different feedback from the user's search direction when the user's design is fixed. Through this study, it can contribute to expanding the way design computing can support designers.