DipGuava: Disentangling Personalized Gaussian Features for 3D Head Avatars from Monocular Video
This work addresses the limitation of capturing personalized details in 3D head avatars for applications like animation and virtual reality, representing a novel method for a known bottleneck.
The paper tackled the problem of 3D head avatar creation from monocular video by introducing DipGuava, a method that disentangles facial appearance into geometry-driven base and personalized residual components, resulting in avatars with improved realism and expressiveness, as shown by outperforming prior methods in visual quality and quantitative performance.
While recent 3D head avatar creation methods attempt to animate facial dynamics, they often fail to capture personalized details, limiting realism and expressiveness. To fill this gap, we present DipGuava (Disentangled and Personalized Gaussian UV Avatar), a novel 3D Gaussian head avatar creation method that successfully generates avatars with personalized attributes from monocular video. DipGuava is the first method to explicitly disentangle facial appearance into two complementary components, trained in a structured two-stage pipeline that significantly reduces learning ambiguity and enhances reconstruction fidelity. In the first stage, we learn a stable geometry-driven base appearance that captures global facial structure and coarse expression-dependent variations. In the second stage, the personalized residual details not captured in the first stage are predicted, including high-frequency components and nonlinearly varying features such as wrinkles and subtle skin deformations. These components are fused via dynamic appearance fusion that integrates residual details after deformation, ensuring spatial and semantic alignment. This disentangled design enables DipGuava to generate photorealistic, identity-preserving avatars, consistently outperforming prior methods in both visual quality and quantitativeperformance, as demonstrated in extensive experiments.