Geometry-Contrastive GAN for Facial Expression Transfer

Abstract

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.

References in corpus (7)

• Conditional Generative Adversarial Nets
• TensorFlow: A system for large-scale machine learning
• Improved Training of Wasserstein GANs
• Age Progression/Regression by Conditional Adversarial Autoencoder
• Towards Large-Pose Face Frontalization in the Wild
• Photorealistic Facial Expression Synthesis by the Conditional Difference Adversarial Autoencoder
• Differential Generative Adversarial Networks: Synthesizing Non-linear Facial Variations with Limited Number of Training Data

Cited by in corpus (14)

• Deep Facial Expression Recognition: A Survey
• Adversarial Training in Affective Computing and Sentiment Analysis: Recent Advances and Perspectives
• Cross-Modal Contrastive Learning for Text-to-Image Generation
• F3A-GAN: Facial Flow for Face Animation with Generative Adversarial Networks
• X2Face: A network for controlling face generation by using images, audio, and pose codes
• Cascade EF-GAN: Progressive Facial Expression Editing with Local Focuses
• GestureGAN for Hand Gesture-to-Gesture Translation in the Wild
• Make a Face: Towards Arbitrary High Fidelity Face Manipulation
• Perceptually Validated Precise Local Editing for Facial Action Units with StyleGAN
• LEED: Label-Free Expression Editing via Disentanglement
• Synthetic Expressions are Better Than Real for Learning to Detect Facial Actions
• Joint Deep Learning of Facial Expression Synthesis and Recognition
• 3D Dense Geometry-Guided Facial Expression Synthesis by Adversarial Learning
• Learn to synthesize and synthesize to learn