Gaussian3Diff: 3D Gaussian Diffusion for 3D Full Head Synthesis and Editing

TL;DR

Gaussian3Diff introduces a novel 3D diffusion framework using Gaussian-based implicit functions and UV parameterization for photorealistic 3D head synthesis and editing, enabling flexible manipulation of facial features and expressions.

Contribution

The paper proposes a new 3D head synthesis method combining Gaussian implicit functions with UV space parameterization, enhancing realism and editing capabilities.

Findings

Produces diverse, realistic 3D head models.

Enables fine-grained facial editing.

Demonstrates effectiveness through extensive experiments.

Abstract

We present a novel framework for generating photorealistic 3D human head and subsequently manipulating and reposing them with remarkable flexibility. The proposed approach leverages an implicit function representation of 3D human heads, employing 3D Gaussians anchored on a parametric face model. To enhance representational capabilities and encode spatial information, we embed a lightweight tri-plane payload within each Gaussian rather than directly storing color and opacity. Additionally, we parameterize the Gaussians in a 2D UV space via a 3DMM, enabling effective utilization of the diffusion model for 3D head avatar generation. Our method facilitates the creation of diverse and realistic 3D human heads with fine-grained editing over facial features and expressions. Extensive experiments demonstrate the effectiveness of our method.