MoGaFace: Momentum-Guided and Texture-Aware Gaussian Avatars for Consistent Facial Geometry

TL;DR

MoGaFace introduces a novel framework for 3D head avatar reconstruction that refines facial geometry and texture continuously, ensuring high-fidelity and consistent avatars even with initial inaccuracies and real-world data.

Contribution

It proposes the Momentum-Guided Consistent Geometry and Latent Texture Attention modules for improved alignment and texture refinement during Gaussian-based avatar rendering.

Findings

Achieves high-fidelity head avatar reconstruction.

Significantly improves novel-view synthesis quality.

Effective under inaccurate mesh initialization and real-world conditions.

Abstract

Existing 3D head avatar reconstruction methods adopt a two-stage process, relying on tracked FLAME meshes derived from facial landmarks, followed by Gaussian-based rendering. However, misalignment between the estimated mesh and target images often leads to suboptimal rendering quality and loss of fine visual details. In this paper, we present MoGaFace, a novel 3D head avatar modeling framework that continuously refines facial geometry and texture attributes throughout the Gaussian rendering process. To address the misalignment between estimated FLAME meshes and target images, we introduce the Momentum-Guided Consistent Geometry module, which incorporates a momentum-updated expression bank and an expression-aware correction mechanism to ensure temporal and multi-view consistency. Additionally, we propose Latent Texture Attention, which encodes compact multi-view features into head-aware representations, enabling geometry-aware texture refinement via integration into Gaussians. Extensive experiments show that MoGaFace achieves high-fidelity head avatar reconstruction and significantly improves novel-view synthesis quality, even under inaccurate mesh initialization and unconstrained real-world settings.