FaceFeat-GAN: a Two-Stage Approach for Identity-Preserving Face Synthesis

TL;DR

FaceFeat-GAN introduces a two-stage generative model that enhances identity preservation, diversity, and image quality in face synthesis by separately generating features and rendering images.

Contribution

It proposes a novel two-stage framework that separates feature generation from image rendering, improving diversity and identity preservation over existing single-stage models.

Findings

Outperforms previous methods in identity preservation and diversity.

Generates highly realistic and diverse face images.

Effectively maintains facial identity across diverse outputs.

Abstract

The advance of Generative Adversarial Networks (GANs) enables realistic face image synthesis. However, synthesizing face images that preserve facial identity as well as have high diversity within each identity remains challenging. To address this problem, we present FaceFeat-GAN, a novel generative model that improves both image quality and diversity by using two stages. Unlike existing single-stage models that map random noise to image directly, our two-stage synthesis includes the first stage of diverse feature generation and the second stage of feature-to-image rendering. The competitions between generators and discriminators are carefully designed in both stages with different objective functions. Specially, in the first stage, they compete in the feature domain to synthesize various facial features rather than images. In the second stage, they compete in the image domain to render photo-realistic images that contain high diversity but preserve identity. Extensive experiments show that FaceFeat-GAN generates images that not only retain identity information but also have high diversity and quality, significantly outperforming previous methods.