Files
Abstract
Sketching is a critical part of the early stages of the design process, facilitating ideation and the exploration of conceptual designs. Digital sketching tools have been introduced as a method for augmenting and supporting the sketching process. In contrast, intelligent systems that collaborate with designers on creative tasks are referred to as computational co-creative systems. These systems contribute to a shared creative artifact with users in a process that can support and inspire user creativity. A common occurrence in sketching creativity is the conceptual shift, or when a drawn object is re-interpreted as belonging to a different object category. Identifying and capitalizing on conceptual shifts is an important part of the creative process as they involve re-interpreting input in a new context, category, or domain.We introduce a co-creative design system called Creative Sketching Partner (CSP), which involves collaboration between a designer and an AI agent on a shared design task. Our AI model for making conceptual shifts can analyze and extract visual and conceptual features from the user’s sketched object and then identify a relevant object to display in order to encourage user creativity. We describe our computational model in identifying and generating conceptual shifts, followed by different scenarios to demonstrate the results of our algorithm in a design context.Our tool is intended to encourage creativity, facilitate creative ideation, and overcome design fixation. In addition, we want our tool to support different forms of creativity, such as combinatorial, exploratory, and transformational, depending on the parameters selected for the operation of the model. We presume that the degree of similarity between the user’s and the system’s sketches is associated with a range of cognitive models of creativity in a design context. We report on the findings of an empirical study that analyzes different design scenarios in which the user sketches in response to a proposed conceptual shift. The findings show that high visual and conceptual similarity is associated with combinatorial creativity and low visual and conceptual similarity is associated with transformational creativity.