cs.GRMar 25, 2026

FaceParts: Segmentation and Editing of Gaussian Splatting

Authors: Tymoteusz Zapała, Julia Farganus, Dominik Galus, Mikołaj Czachorowski, Piotr Syga, Przemysław Spurek

Organizations: Wrocław University of Science and Technology, Poland · Jagiellonian University, Poland

Read PDF
Ask AI
Cite

Copy citation

DOINot available for this paper

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).

Figures & tables

Figure 1: Overview of the FaceParts reconstruction pipeline. The unsupervised segmentation process has three stages: Disentanglement , where a neural network groups similar Gaussians into shared facial components; Refinement , where density-based clustering extracts coherent subparts; and Extraction , where these parts are stored for reuse. The framework supports facial part swapping , enabling modification of a target avatar by transferring features (e.g., beard, eyes).
Figure 2: Framework overview: (a) Head avatars employ a hybrid solution of 3D Gaussians rigged to a parametric mesh model. (b) Reconstructing Gaussian features based solely on position encourages grouping of similar semantic regions within the same bottleneck channel. (c) Density-based clustering ensures that the components constitute smaller logical facial parts, such as an eyebrow. (d) The underlying FLAME model enables transferring facial parts without manual adjustments.
Figure 3: Examples of face-part transfer across identities. The top row shows extracted segments (beard, eyebrows, eyes, and mustache) from avatar 218, while the bottom row shows the same parts attached to avatar 210.
Figure 4: Dynamic behavior of face swapping. Top row: avatar 253 rendered across different poses and expressions. Bottom row: the same avatar with a beard transferred from avatar 218. The transferred segment adapts consistently to head pose and expression, following the deformations of the underlying FLAME mesh.
Figure 5: Segmentation results for different bottleneck sizes. Smaller bottlenecks (e.g., size 3 and 4) lead to interpretable groupings such as clothing, skin, and other features, while larger bottlenecks often produce fragmented clusters without clear semantic meaning.
Figure 6: Example highlighting the limitations of segmentation and generative models. (a) Segmentation, left: ground-truth image. Middle: output from a 2D face segmentation model. Right: our result. (b) Editing, middle: inpainting with a generative diffusion model. Right: our result. The segmentation model struggles to capture realistic appearance and fine-grained details, while the diffusion model lacks reliability in out-of-distribution cases (e.g., woman with a beard). Our approach preserves identity and structure while producing natural and consistent facial features.
SegmentAED ↓\downarrowAPD ↓\downarrowID ↑\uparrowFID ↓\downarrow
Beard0.037 (0.015)0.007 (0.006)0.905 (0.011)–
Eyebrows0.014 (0.006)0.004 (0.003)0.927 (0.007)–
Eyes0.004 (0.002)0.001 (0.001)0.986 (0.003)–
Mustache0.029 (0.014)0.005 (0.004)0.952 (0.010)–
Global Avg0.021 (0.009)0.004 (0.003)0.943 (0.008)5.872
Table 1: Quantitative evaluation of face-part transfer. Metrics are averaged across all avatars and grouped by segment type. Results indicate that attaching an additional segment has only a small influence on expression and pose extraction. The largest variations are observed for the beard, likely because it is the largest segment. The overall FID score is also reported to quantify realism relative to unmodified avatars.
Figure 7: Comparison of transferring hair from avatar 306 to avatar 218 using our FaceParts and MEGA ( Wang et al., 2024a ) .
Figure 8: a) Lusage+Lsparsity\mathcal{L}_{usage}+\mathcal{L}_{sparsity} b) Lsparsity\mathcal{L}_{sparsity} c) Lusage\mathcal{L}_{usage} d) no additional loss e) direct prediction on xyzxyz coordinates f) prediction on xyzxyz with the standard softmax. The sparsity loss encourages the network to cluster nearby regions, while the usage loss promotes a more uniform bottleneck distribution across the entire avatar. Models trained without HashGrid fail to disentangle smaller facial parts.
Appendix figures & tables3 assets

Supplementary material from the paper’s appendix.

Appendix

Figure 9: The first row illustrates the overlap strategy with different values of parameter oo . Higher values make the attached beard less visible, as the avatar’s Gaussians dominate. The second row shows the replacement strategy, where smaller values of parameter nn result in a more visible beard.
Figure 10: Examples of facial part swaps.
Figure 11: First row: hair extracted from avatar 306 using the disentanglement network followed by clustering. Second row: face swap with an additional step. Beyond transferring Gaussians in relative coordinates, we locally adjust the FLAME mesh for avatar 218 (only the triangles corresponding to the hair region of avatar 306) for better alignment.
More in this topic

3DGS Segmentation

See all 15 papers

Explore similar work

CardsList

  1. Edit3DGS: Unified Framework for Dynamic Head Editing via 2D Instruction-Guided Diffusion and 3D Gaussian Splatting

    Jun 16, 2026Duy-Dat Tran, Trung-Nghia Le3D Scene EditingVideo Editing

    University of Science, VNU-HCM, Ho Chi Minh, Vietnam · Vietnam National University, Ho Chi Minh, Vietnam

  2. ManifoldSplat: Language-Guided Semantic Shape Editing of 3D Gaussian Head Avatars

    Oct 2, 2026Antonio Canela, Jordi Sànchez-Riera3D Shape Editing3D Gaussian Splatting

    Universitat Politècnica de Catalunya · Institut de Robòtica i Informàtica Industrial, CSIC-UPC

  3. ChromaGS: Text-Driven Semantic Editing of 4D Gaussian Avatars

    Oct 2, 2026Antonio Canela, Jordi Sànchez-Riera3D Scene Editing3D Gaussian Splatting

    Universitat Polit`ecnica de Catalunya · Institut de Rob`otica i Inform`atica Industrial, CSIC-UPC