Audio-Driven Talking Face Generation with Blink Embedding and Hash Grid Landmarks Encoding

TL;DR

This paper introduces a novel method for generating high-fidelity talking face videos by combining blink embedding, hash grid landmarks encoding, and neural radiance fields to improve mouth movement accuracy and realism.

Contribution

It presents an automatic approach that integrates facial and audio features using a Dynamic Landmark Transformer to enhance talking face generation quality.

Findings

Outperforms existing methods in fidelity and realism.

Effectively captures mouth movements and facial expressions.

Produces lifelike 3D talking portraits.

Abstract

Dynamic Neural Radiance Fields (NeRF) have demonstrated considerable success in generating high-fidelity 3D models of talking portraits. Despite significant advancements in the rendering speed and generation quality, challenges persist in accurately and efficiently capturing mouth movements in talking portraits. To tackle this challenge, we propose an automatic method based on blink embedding and hash grid landmarks encoding in this study, which can substantially enhance the fidelity of talking faces. Specifically, we leverage facial features encoded as conditional features and integrate audio features as residual terms into our model through a Dynamic Landmark Transformer. Furthermore, we employ neural radiance fields to model the entire face, resulting in a lifelike face representation. Experimental evaluations have validated the superiority of our approach to existing methods.