Deformable Style Transfer
This work solves the limitation of existing style transfer methods that focus only on texture, making it applicable to a broader range of images without domain restrictions.
The paper tackles the problem of style transfer by addressing both texture and geometry, which previous methods largely ignored, and demonstrates the approach on diverse image types.
Both geometry and texture are fundamental aspects of visual style. Existing style transfer methods, however, primarily focus on texture, almost entirely ignoring geometry. We propose deformable style transfer (DST), an optimization-based approach that jointly stylizes the texture and geometry of a content image to better match a style image. Unlike previous geometry-aware stylization methods, our approach is neither restricted to a particular domain (such as human faces), nor does it require training sets of matching style/content pairs. We demonstrate our method on a diverse set of content and style images including portraits, animals, objects, scenes, and paintings. Code has been made publicly available at https://github.com/sunniesuhyoung/DST.