The #paperoftheweek 40 was Geometry-Contrastive Generative Adversarial Network for Facial Expression Synthesis.

For this week we found an interesting paper that focuses on the transfer of continuous emotions across different subjects. A conditional GAN is applied to facial expression transfer. Contrastive learning helps to embed facial geometry into a semantic manifold of facial expressions.

The authors of this paper 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 their proposed method can be applied in facial expression transfer even there exist big differences in facial shapes and expressions between different subjects.

You can find the full paper here: https://arxiv.org/abs/1802.01822