Abstract
Facial attribute manipulation has attracted great attention from the public due to its wide range of applications. Aiming to smoothly manipulate the attributes of real facial images, it is critical to search a proper latent code aligns with the domain of pre-trained GAN for faithful inversion and control the transformation within the scope of the attribute for precise editing. Previous methods mainly focused on improving the quality of reconstruction, but often ignored the editing effect. To address this issue, we first propose a mapping network to manipulate latent code which is effective for diverse situations, and design a spatial attention network to predict binary mask of certain attribute which encourages to only alter relevant region of images and suppress irrelevant changes. In addition, we introduce a novel latent space into GAN inversion framework which achieves high reconstruction quality especially preserving identity features and retains ability to edit face attributes. Our methods pave the way to semantically meaningful and disentangled manipulations on both generated images and real images. Experimental results indicate a clear improvement over the current state-of-the-art methods both in subjective and objective metrics.
