WEM-GAN: Wavelet transform based facial expression manipulation

TL;DR

WEM-GAN is a novel wavelet transform-based GAN that enhances facial expression manipulation by preserving facial details and identity, outperforming previous methods in quality and accuracy on the AffectNet dataset.

Contribution

The paper introduces WEM-GAN, which integrates wavelet transforms and high-frequency discriminators to improve detail preservation and expression editing in facial manipulation.

Findings

Better preservation of facial identity features.

Superior performance in expression editing accuracy.

Improved image quality demonstrated on AffectNet dataset.

Abstract

Facial expression manipulation aims to change human facial expressions without affecting face recognition. In order to transform the facial expressions to target expressions, previous methods relied on expression labels to guide the manipulation process. However, these methods failed to preserve the details of facial features, which causes the weakening or the loss of identity information in the output image. In our work, we propose WEM-GAN, in short for wavelet-based expression manipulation GAN, which puts more efforts on preserving the details of the original image in the editing process. Firstly, we take advantage of the wavelet transform technique and combine it with our generator with a U-net autoencoder backbone, in order to improve the generator's ability to preserve more details of facial features. Secondly, we also implement the high-frequency component discriminator, and use high-frequency domain adversarial loss to further constrain the optimization of our model, providing the generated face image with more abundant details. Additionally, in order to narrow the gap between generated facial expressions and target expressions, we use residual connections between encoder and decoder, while also using relative action units (AUs) several times. Extensive qualitative and quantitative experiments have demonstrated that our model performs better in preserving identity features, editing capability, and image generation quality on the AffectNet dataset. It also shows superior performance in metrics such as Average Content Distance (ACD) and Expression Distance (ED).