Learning Perceptual Aesthetics of 3-D Shapes From Multiple Views.

Abstract

The quantification of 3-D shape aesthetics has so far focused on specific shape features and manually defined criteria such as the curvature and the rule of thirds. In this article, we built a model of 3-D shape aesthetics directly from human aesthetics preference data and show it to be well aligned with human perception of aesthetics. To build this model, we first crowdsource a large number of human aesthetics preferences by showing shapes in pairs in an online study and then use the same to build a 3-D shape multiview-based deep neural network architecture to allow us to learn a measure of 3-D shape aesthetics. In comparison to previous approaches, we do not use any predefined notions of aesthetics to build our model. Our algorithmically computed measure of shape aesthetics is beneficial to a range of applications in graphics such as search, visualization, and scene composition.