Make a Face: Towards Arbitrary High Fidelity Face Manipulation

TL;DR

This paper introduces AF-VAE, a novel high-resolution face manipulation method that achieves high fidelity and diversity using weak supervision, an additive Gaussian mixture assumption, and perceptual quality improvements inspired by the Human Visual System.

Contribution

The paper proposes AF-VAE with an additive Gaussian mixture model and HVS-inspired architecture strategies, enabling high-resolution, diverse, and high-quality face manipulation with weak supervision.

Findings

Achieves state-of-the-art Inception Score and FID scores.

Demonstrates superior face manipulation fidelity and extremity.

Provides fine control over sample quality and model complexity.

Abstract

Recent studies have shown remarkable success in face manipulation task with the advance of GANs and VAEs paradigms, but the outputs are sometimes limited to low-resolution and lack of diversity. In this work, we propose Additive Focal Variational Auto-encoder (AF-VAE), a novel approach that can arbitrarily manipulate high-resolution face images using a simple yet effective model and only weak supervision of reconstruction and KL divergence losses. First, a novel additive Gaussian Mixture assumption is introduced with an unsupervised clustering mechanism in the structural latent space, which endows better disentanglement and boosts multi-modal representation with external memory. Second, to improve the perceptual quality of synthesized results, two simple strategies in architecture design are further tailored and discussed on the behavior of Human Visual System (HVS) for the first time, allowing for fine control over the model complexity and sample quality. Human opinion studies and new state-of-the-art Inception Score (IS) / Frechet Inception Distance (FID) demonstrate the superiority of our approach over existing algorithms, advancing both the fidelity and extremity of face manipulation task.