Abstract

Geometry and shape are fundamental aspects of visual style. Existing styletransfer methods focus on texture-like components of style, ignoring geometry.We propose deformable style transfer (DST), an optimization-based approach thatintegrates texture and geometry style transfer. Our method is the first toallow geometry-aware stylization not restricted to any domain and not requiringtraining sets of matching style/content pairs. We demonstrate our method on adiverse set of content and style images including portraits, animals, objects,scenes, and paintings.

 

Quick Read (beta)