GeoDiff4D: Geometry-Aware Diffusion for 4D Head Avatar Reconstruction

TL;DR

GeoDiff4D introduces a geometry-aware diffusion framework that synthesizes high-fidelity 4D head avatars from a single image, effectively capturing geometry and expression for realistic, animatable reconstructions.

Contribution

The paper presents a novel geometry-aware diffusion method that jointly synthesizes images and surface normals, improving 3D consistency and realism in 4D head avatar reconstruction from a single portrait.

Findings

Outperforms state-of-the-art methods in visual quality and expression fidelity

Supports real-time rendering of 4D head avatars

Achieves better cross-identity generalization

Abstract

Reconstructing photorealistic and animatable 4D head avatars from a single portrait image remains a fundamental challenge in computer vision. While diffusion models have enabled remarkable progress in image and video generation for avatar reconstruction, existing methods primarily rely on 2D priors and struggle to achieve consistent 3D geometry. We propose a novel framework that leverages geometry-aware diffusion to learn strong geometry priors for high-fidelity head avatar reconstruction. Our approach jointly synthesizes portrait images and corresponding surface normals, while a pose-free expression encoder captures implicit expression representations. Both synthesized images and expression latents are incorporated into 3D Gaussian-based avatars, enabling photorealistic rendering with accurate geometry. Extensive experiments demonstrate that our method substantially outperforms state-of-the-art approaches in visual quality, expression fidelity, and cross-identity generalization, while supporting real-time rendering.