QUANTIFYING SHAPE DESCRIPTORS FOR AESTHETIC CONCEPTS
A new approach to quantitatively estimate the degree of perception of aesthetic concepts for object silhouettes is proposed. First, a survey has been conducted to obtain statistical measurements of the perception degree of the interviewees with respect to the opposite concept pair Gentle-Aggressive for a set of object silhouettes. Then, a new shape descriptor and the statistical measurement, associated to each silhouette, are utilized to train an artificial neural network. The obtained model allows predicting the perception degree of the studied aesthetic concepts for new silhouettes.