DisentTalk: Cross-lingual Talking Face Generation via Semantic Disentangled Diffusion Model
This work addresses the challenge of cross-lingual talking face generation for applications like animation and virtual avatars, offering an incremental improvement by integrating 3DMM and diffusion-based approaches.
The paper tackles the problem of generating talking faces with fine-grained control and temporal consistency by introducing DisentTalk, a semantic disentanglement framework that decomposes 3DMM expression parameters and uses a hierarchical latent diffusion model, achieving superior performance in lip synchronization, expression quality, and temporal consistency over existing methods.
Recent advances in talking face generation have significantly improved facial animation synthesis. However, existing approaches face fundamental limitations: 3DMM-based methods maintain temporal consistency but lack fine-grained regional control, while Stable Diffusion-based methods enable spatial manipulation but suffer from temporal inconsistencies. The integration of these approaches is hindered by incompatible control mechanisms and semantic entanglement of facial representations. This paper presents DisentTalk, introducing a data-driven semantic disentanglement framework that decomposes 3DMM expression parameters into meaningful subspaces for fine-grained facial control. Building upon this disentangled representation, we develop a hierarchical latent diffusion architecture that operates in 3DMM parameter space, integrating region-aware attention mechanisms to ensure both spatial precision and temporal coherence. To address the scarcity of high-quality Chinese training data, we introduce CHDTF, a Chinese high-definition talking face dataset. Extensive experiments show superior performance over existing methods across multiple metrics, including lip synchronization, expression quality, and temporal consistency. Project Page: https://kangweiiliu.github.io/DisentTalk.