Geometry-Contrastive GAN for Facial Expression Transfer
This work addresses facial expression transfer for applications like animation or human-computer interaction, but it appears incremental as it builds on existing cGANs with added geometry and contrastive learning components.
The paper tackles the problem of transferring continuous emotions across different subjects in facial expression transfer by proposing a Geometry-Contrastive GAN (GC-GAN), which uses geometry information and contrastive learning to generate identity-preserving faces with target expressions, achieving effective results even with significant differences in facial shapes and expressions.
In this paper, we propose a Geometry-Contrastive Generative Adversarial Network (GC-GAN) for transferring continuous emotions across different subjects. Given an input face with certain emotion and a target facial expression from another subject, GC-GAN can generate an identity-preserving face with the target expression. Geometry information is introduced into cGANs as continuous conditions to guide the generation of facial expressions. In order to handle the misalignment across different subjects or emotions, contrastive learning is used to transform geometry manifold into an embedded semantic manifold of facial expressions. Therefore, the embedded geometry is injected into the latent space of GANs and control the emotion generation effectively. Experimental results demonstrate that our proposed method can be applied in facial expression transfer even there exist big differences in facial shapes and expressions between different subjects.