Learning to Caricature via Semantic Shape Transform

Abstract Caricature is an artistic drawing created to abstract or exaggerate facial features of a person. Rendering visually pleasing caricatures difficult task that requires professional skills, and thus it great interest design method automatically generate such drawings. To deal with large shape changes, we propose algorithm based on semantic transform produce diverse plausible exaggerations. Specifically, predict pixel-wise correspondences perform image warping the input photo achieve dense transformation. We show proposed framework able render exaggerations while maintaining their structures. In addition, our model allows users manipulate via map. demonstrate effectiveness approach photograph-caricature benchmark dataset comparisons state-of-the-art methods.