StyleCariGAN

We present a caricature generation framework based on shape and style manipulation using StyleGAN. Our framework, dubbed StyleCariGAN , automatically creates realistic detailed from an input photo with optional controls exaggeration degree color stylization type. The key component of our method is blocks that are used for modulating coarse layer feature maps StyleGAN to produce desirable exaggerations. first build layer-mixed photo-to-caricature conversion by swapping fine layers the photos corresponding trained generate caricatures. Given photo, model produces but without then append train create exaggerations while preserving characteristic appearances input. Experimental results show generates caricatures compared current state-of-the-art methods. demonstrate also supports other StyleGAN-based image manipulations, such as facial expression control.