
GaussianTalker is a novel framework for real-time generation of pose-controllable talking heads by leveraging the fast rendering capabilities of 3D Gaussian Splatting (3DGS) and addressing the challenges of directly controlling 3DGS with speech audio.


coremsg

real-time-high-fidelity-talking-head-synthesis-with-audio-driven-3d-gaussian-splatting


Real-Time High-Fidelity Talking Head Synthesis with Audio-Driven 3D Gaussian Splatting


title_rewrite


A hierarchical diffusion framework, DreamHead, is proposed to effectively learn the spatial-temporal correspondence between audio input and facial dynamics for high-quality talking head video synthesis.


hierarchical-diffusion-for-accurate-and-smooth-audio-driven-talking-head-synthesis


Hierarchical Diffusion for Accurate and Smooth Audio-driven Talking Head Synthesis
