Stereo-Talker: Audio-driven 3D Human Synthesis with Prior-Guided Mixture-of-Experts
This addresses the problem of creating realistic and controllable talking human videos for applications like virtual avatars or entertainment, though it builds incrementally on existing audio-driven synthesis methods.
The paper tackles audio-driven 3D human video synthesis by introducing Stereo-Talker, a system that generates talking videos with precise lip sync, expressive gestures, and viewpoint control, using a two-stage approach with LLM priors and a Mixture-of-Experts mechanism, and it includes a dataset of 2,203 identities for broad generalization.
This paper introduces Stereo-Talker, a novel one-shot audio-driven human video synthesis system that generates 3D talking videos with precise lip synchronization, expressive body gestures, temporally consistent photo-realistic quality, and continuous viewpoint control. The process follows a two-stage approach. In the first stage, the system maps audio input to high-fidelity motion sequences, encompassing upper-body gestures and facial expressions. To enrich motion diversity and authenticity, large language model (LLM) priors are integrated with text-aligned semantic audio features, leveraging LLMs' cross-modal generalization power to enhance motion quality. In the second stage, we improve diffusion-based video generation models by incorporating a prior-guided Mixture-of-Experts (MoE) mechanism: a view-guided MoE focuses on view-specific attributes, while a mask-guided MoE enhances region-based rendering stability. Additionally, a mask prediction module is devised to derive human masks from motion data, enhancing the stability and accuracy of masks and enabling mask guiding during inference. We also introduce a comprehensive human video dataset with 2,203 identities, covering diverse body gestures and detailed annotations, facilitating broad generalization. The code, data, and pre-trained models will be released for research purposes.