Abstraction of 2D shapes in terms of parts

Abstraction in imagery results from the strategic simplification and elimination of detail to clarify the visual structure of the depicted shape. It is a mainstay of artistic practice and an important ingredient of effective visual communication. We develop a computational method for the abstract depiction of 2D shapes. Our approach works by organizing the shape into parts using a new synthesis of holistic features of the part shape, local features of the shape boundary, and global aspects of shape organization. Our abstractions are new shapes with fewer and clearer parts.