Arbitrary Talking Face Generation via Attentional Audio-Visual Coherence Learning

Abstract

Talking face generation aims to synthesize a face video with precise lip synchronization as well as a smooth transition of facial motion over the entire video via the given speech clip and facial image. Most existing methods mainly focus on either disentangling the information in a single image or learning temporal information between frames. However, cross-modality coherence between audio and video information has not been well addressed during synthesis. In this paper, we propose a novel arbitrary talking face generation framework by discovering the audio-visual coherence via the proposed Asymmetric Mutual Information Estimator (AMIE). In addition, we propose a Dynamic Attention (DA) block by selectively focusing the lip area of the input image during the training stage, to further enhance lip synchronization. Experimental results on benchmark LRW dataset and GRID dataset transcend the state-of-the-art methods on prevalent metrics with robust high-resolution synthesizing on gender and pose variations.

IJCAI-2020

References in corpus (8)

• Estimating Mutual Information
• Learning deep representations by mutual information estimation and maximization
• Generating Videos with Scene Dynamics
• Which Training Methods for GANs do actually Converge?
• Mutual information for the selection of relevant variables in spectrometric nonlinear modelling
• ObamaNet: Photo-realistic lip-sync from text
• You said that?
• Speech-Driven Facial Reenactment Using Conditional Generative Adversarial Networks

Cited by in corpus (7)

• A Review on Generative Adversarial Networks: Algorithms, Theory, and Applications
• Deep Audio-Visual Learning: A Survey
• Mutual Information Maximization for Effective Lip Reading
• A Survey on Audio Synthesis and Audio-Visual Multimodal Processing
• Lets Play Music: Audio-driven Performance Video Generation
• Facial Keypoint Sequence Generation from Audio
• A Comprehensive Multi-scale Approach for Speech and Dynamics Synchrony in Talking Head Generation