Evaluating the Temporal Effect of Image Features on Dynamic Aesthetic Experience of Generative art

Abstract

The emergence of dynamic digital arts, such as dynamic generative art, has reshaped how aesthetic experiences can be studied, emphasizing their inherently dynamic and evolving nature. Within the framework of computational aesthetics, which seeks to model and quantify human perceptions of beauty, this study extends the focus from static to dynamic stimuli. We investigate the temporal relationship between evolving image features and aesthetic judgments, exploring how dynamic visual properties influence the progression of aesthetic judgments over time. By introducing temporal offsets in the analysis and employing linear and nonlinear statistical models, we examine continuous aesthetic ratings alongside an extensive set of image features. Our findings offer a novel perspective on the enduring impact of image features on aesthetic experience.