Neural Face Editing with Intrinsic Image Disentangling
This addresses the need for more efficient and robust face editing tools for users in computer vision and graphics, though it is incremental as it builds on existing disentanglement methods.
The paper tackles the problem of tedious and fragile face editing by proposing an end-to-end GAN that infers a disentangled representation of intrinsic face properties, enabling semantically relevant edits while keeping other aspects fixed.
Traditional face editing methods often require a number of sophisticated and task specific algorithms to be applied one after the other --- a process that is tedious, fragile, and computationally intensive. In this paper, we propose an end-to-end generative adversarial network that infers a face-specific disentangled representation of intrinsic face properties, including shape (i.e. normals), albedo, and lighting, and an alpha matte. We show that this network can be trained on "in-the-wild" images by incorporating an in-network physically-based image formation module and appropriate loss functions. Our disentangling latent representation allows for semantically relevant edits, where one aspect of facial appearance can be manipulated while keeping orthogonal properties fixed, and we demonstrate its use for a number of facial editing applications.