FaceParts: Segmentation and Editing of Gaussian Splatting
For 3D avatar editing, this provides a fully automatic method without manual labor or 2D/3D supervision, though limited to Gaussian Splatting representations.
FaceParts enables unsupervised segmentation and editing of 3D Gaussian Splatting avatars into facial parts, achieving identity consistency (ID=0.943) and low expression/pose distances (AED=0.021, APD=0.004) on the NeRSemble dataset.
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).