Mapping aesthetic properties to 3D free form shapes through the use of a machine learning based framework

Current production is moving from the mass production concept to the product customization and personalization. Customers are not anymore only buyers. Not only they are becoming actors within the Product Development Process (PDP) but, thanks to new production technologies like 3D printers, they can be both designers and producers. In this scenario, the development of user-friendly design tools is crucial. Declarative approaches are suitable and can address such requirements. They exploit generally understood and shared concepts closer to the way people perceive shapes than to the way shapes are modeled with complex geometric models. To this aim, this paper presents a generic framework for understanding the shape characteristics associated to perceptual/aesthetic properties of 3D free form shapes. This framework is used to investigate whether there is a common judgment to characterize the flatness of surfaces and which are the surface shape characteristics affecting the flatness perception? From the experiments, it results that the size and transition of the surrounding influence the perception of the flatness of a given surface strengthening the classification consistency.