GSTalker: Real-time Audio-Driven Talking Face Generation via Deformable Gaussian Splatting

TL;DR

GSTalker is a novel 3D audio-driven talking face generation model that achieves fast training and real-time rendering by using Gaussian Splatting and deformation fields to synchronize facial movements with audio.

Contribution

The paper introduces GSTalker, which employs Gaussian Splatting and deformation fields for efficient, high-fidelity, audio-synchronized 3D talking face generation with significantly reduced training and rendering times.

Findings

Fast training within 40 minutes

Real-time rendering at 125 FPS

High-fidelity, audio-synchronized face generation

Abstract

We present GStalker, a 3D audio-driven talking face generation model with Gaussian Splatting for both fast training (40 minutes) and real-time rendering (125 FPS) with a 3$\sim$5 minute video for training material, in comparison with previous 2D and 3D NeRF-based modeling frameworks which require hours of training and seconds of rendering per frame. Specifically, GSTalker learns an audio-driven Gaussian deformation field to translate and transform 3D Gaussians to synchronize with audio information, in which multi-resolution hashing grid-based tri-plane and temporal smooth module are incorporated to learn accurate deformation for fine-grained facial details. In addition, a pose-conditioned deformation field is designed to model the stabilized torso. To enable efficient optimization of the condition Gaussian deformation field, we initialize 3D Gaussians by learning a coarse static Gaussian representation. Extensive experiments in person-specific videos with audio tracks validate that GSTalker can generate high-fidelity and audio-lips synchronized results with fast training and real-time rendering speed.