Advancing Facial Stylization through Semantic Preservation Constraint and Pseudo-Paired Supervision
This work addresses the challenge of maintaining content consistency in facial stylization for applications in digital art and media, representing an incremental improvement over existing StyleGAN-based approaches.
The paper tackles the problem of facial stylization by addressing semantic shifts and content fidelity, proposing a method with semantic preservation constraint and pseudo-paired supervision that achieves high-fidelity and aesthetically pleasing results surpassing previous methods.
Facial stylization aims to transform facial images into appealing, high-quality stylized portraits, with the critical challenge of accurately learning the target style while maintaining content consistency with the original image. Although previous StyleGAN-based methods have made significant advancements, the generated results still suffer from artifacts or insufficient fidelity to the source image. We argue that these issues stem from neglecting semantic shift of the generator during stylization. Therefore, we propose a facial stylization method that integrates semantic preservation constraint and pseudo-paired supervision to enhance the content correspondence and improve the stylization effect. Additionally, we develop a methodology for creating multi-level pseudo-paired datasets to implement supervisory constraint. Furthermore, building upon our facial stylization framework, we achieve more flexible multimodal and reference-guided stylization without complex network architecture designs or additional training. Experimental results demonstrate that our approach produces high-fidelity, aesthetically pleasing facial style transfer that surpasses previous methods.