CA-GAN: Weakly Supervised Color Aware GAN for Controllable Makeup Transfer

TL;DR

CA-GAN is a weakly supervised generative model that enables continuous and controllable makeup color transfer on specific facial objects, addressing the lack of explicit control in existing methods.

Contribution

The paper introduces CA-GAN, a novel weakly supervised GAN that allows explicit control over makeup color transfer for specific facial regions.

Findings

Achieves controllable makeup color transfer with high fidelity.

Requires only weak attribute labels for training.

Provides the first quantitative analysis of makeup style transfer.

Abstract

While existing makeup style transfer models perform an image synthesis whose results cannot be explicitly controlled, the ability to modify makeup color continuously is a desirable property for virtual try-on applications. We propose a new formulation for the makeup style transfer task, with the objective to learn a color controllable makeup style synthesis. We introduce CA-GAN, a generative model that learns to modify the color of specific objects (e.g. lips or eyes) in the image to an arbitrary target color while preserving background. Since color labels are rare and costly to acquire, our method leverages weakly supervised learning for conditional GANs. This enables to learn a controllable synthesis of complex objects, and only requires a weak proxy of the image attribute that we desire to modify. Finally, we present for the first time a quantitative analysis of makeup style transfer and color control performance.