FaceParts: Segmentation and Editing of Gaussian Splatting

TL;DR

FaceParts introduces an unsupervised framework for segmenting and editing Gaussian Splatting avatars directly in the Gaussian domain, enabling precise facial part manipulation without manual labeling.

Contribution

The paper presents a novel method for unsupervised facial segmentation and editing in 3D Gaussian avatars, bypassing traditional 2D or mesh-based approaches.

Findings

Robust isolation of facial features like beards, eyebrows, eyes, and mustaches.

Transferred segments adapt well to pose and expression changes.

Maintains high identity consistency during editing.

Abstract

Facial editing is an important task with applications in entertainment, virtual reality, and digital avatars. Most existing approaches rely on generative models in the 2D image domain, while in 3D the task is typically performed through labor-intensive manual editing. We propose FaceParts, a framework for unsupervised segmentation and editing of Gaussian Splatting avatars. Unlike existing 2D or mesh-assisted methods, our approach operates directly in the Gaussian domain, decomposing avatars into semantically coherent facial parts without supervision. The method integrates feature disentanglement, density-based clustering, and FLAME-anchored part transfer, enabling precise editing and cross-avatar part swapping. Experiments on the NeRSemble dataset with 11 subjects demonstrate robust isolation of features such as beards, eyebrows, eyes and mustaches. Quantitative evaluation confirms that transferred segments adapt to pose and expression, while maintaining identity consistency (ID = 0.943), low Average Expression Distance (AED = 0.021) and low Average Pose Distance (APD = 0.004).