Modeling aesthetic preferences: Color coordination and fuzzy sets

Abstract We are exposed to various aesthetic experiences everyday. Important variables involved in aesthetics are preference and harmony. Preference for color schemes can be predicted by combining preferences for the basic colors and ratings of color harmony. We use FHSI color model, in which color channels distributions are expressed with fuzzy membership functions. Such phenomena, as preference and harmony, are usually approximate and thus categorizing and ordering them is a challenge. Using fuzzy linguistic variables and hedges, we offer a human-consistent classification of aesthetic judgements. The main output of the model is aesthetic judgement on a harmony and personal preference level, e.g. very high preference. For example, in the context of apparel coordination, it allows predicting a preference for a look based on clothing colors. According to experiments, the model results in useful predictions of ratings of aesthetic pleasantness. Aesthetic preference prediction has potential in a wide range of domains, including e-commerce, design, marketing, and art.