MyStyle++: A Controllable Personalized Generative Prior
SIGGRAPH Asia 2023

Abstract

overview

In this paper, we propose an approach to obtain a personalized generative prior with explicit control over a set of attributes. We build upon MyStyle, a recently introduced method, that tunes the weights of a pre-trained StyleGAN face generator on a few images of an individual. This system allows synthesizing, editing, and enhancing images of the target individual with high fidelity to their facial features. However, MyStyle does not demonstrate precise control over the attributes of the generated images. We propose to address this problem through a novel optimization system that organizes the latent space in addition to tuning the generator. Our key contribution is to formulate a loss that arranges the latent codes, corresponding to the input images, along a set of specific directions according to their attributes. We demonstrate that our approach, dubbed MyStyle++, is able to synthesize, edit, and enhance images of an individual with great control over the attributes, while preserving the unique facial characteristics of that individual.

Supplementary Video

BibTeX 

                @article{Zeng_2023_mystyle++,
                    author = {Zeng, Libing and Chen, Lele and Xu, Yi and Kalantari, Nima Khademi},
                    title = {MyStyle++: A Controllable Personalized Generative Prior},
                    booktitle={ACM SIGGRAPH Asia},
                    year={2023}
                }

Acknowledgements

We express our gratitude to the anonymous reviewers for their insightful comments and suggestions. Additionally, we would like to thank Keqiang Yan and Yongqing Liang for the valuable discussions. The website template was borrowed from Michael Gharbi.