GeoDiff4D: Geometry-Aware Diffusion for 4D Head Avatar Reconstruction
This work addresses the problem of creating realistic and animatable digital head avatars for applications in computer vision and graphics, representing a novel method for a known bottleneck in avatar reconstruction.
The paper tackles the challenge of reconstructing photorealistic and animatable 4D head avatars from a single portrait image by proposing a geometry-aware diffusion framework that jointly synthesizes images and surface normals, resulting in substantial improvements in visual quality, expression fidelity, and cross-identity generalization over state-of-the-art methods.
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.