DiffusionAvatars: Deferred Diffusion for High-fidelity 3D Head Avatars

TL;DR

DiffusionAvatars introduces a novel diffusion-based neural rendering method for creating high-fidelity, controllable 3D head avatars that maintain temporal consistency across poses and expressions.

Contribution

The paper presents a deferred diffusion approach leveraging a neural parametric head model and cross-attention conditioning to improve 3D head avatar synthesis.

Findings

Produces high-quality, temporally consistent 3D head avatars

Outperforms existing methods in visual quality and consistency

Effective in self-reenactment and animation scenarios

Abstract

DiffusionAvatars synthesizes a high-fidelity 3D head avatar of a person, offering intuitive control over both pose and expression. We propose a diffusion-based neural renderer that leverages generic 2D priors to produce compelling images of faces. For coarse guidance of the expression and head pose, we render a neural parametric head model (NPHM) from the target viewpoint, which acts as a proxy geometry of the person. Additionally, to enhance the modeling of intricate facial expressions, we condition DiffusionAvatars directly on the expression codes obtained from NPHM via cross-attention. Finally, to synthesize consistent surface details across different viewpoints and expressions, we rig learnable spatial features to the head's surface via TriPlane lookup in NPHM's canonical space. We train DiffusionAvatars on RGB videos and corresponding fitted NPHM meshes of a person and test the obtained avatars in both self-reenactment and animation scenarios. Our experiments demonstrate that DiffusionAvatars generates temporally consistent and visually appealing videos for novel poses and expressions of a person, outperforming existing approaches.