Face Swapping as A Simple Arithmetic Operation

TL;DR

This paper introduces Arithmetic Face Swapping (AFS), a novel method that explicitly disentangles identity and style in a pretrained StyleGAN's latent space, enabling simple arithmetic operations for high-fidelity face swapping and other manipulations.

Contribution

The paper presents a new disentanglement approach in StyleGAN's latent space that simplifies face swapping as an arithmetic operation and supports versatile face editing tasks.

Findings

AFS outperforms state-of-the-art face swapping methods in quality.

The method guarantees identity preservation through a theoretical condition.

Supports combining identities and styles across multiple faces.

Abstract

We propose a novel high-fidelity face swapping method called "Arithmetic Face Swapping" (AFS) that explicitly disentangles the intermediate latent space W+ of a pretrained StyleGAN into the "identity" and "style" subspaces so that a latent code in W+ is the sum of an "identity" code and a "style" code in the corresponding subspaces. Via our disentanglement, face swapping (FS) can be regarded as a simple arithmetic operation in W+, i.e., the summation of a source "identity" code and a target "style" code. This makes AFS more intuitive and elegant than other FS methods. In addition, our method can generalize over the standard face swapping to support other interesting operations, e.g., combining the identity of one source with styles of multiple targets and vice versa. We implement our identity-style disentanglement by learning a neural network that maps a latent code to a "style" code. We provide a condition for this network which theoretically guarantees identity preservation of the source face even after a sequence of face swapping operations. Extensive experiments demonstrate the advantage of our method over state-of-the-art FS methods in producing high-quality swapped faces. Our source code was made public at https://github.com/truongvu2000nd/AFS