mlsanityData to 14 Oct 2026

GeoAvatar: Adaptive Geometrical Gaussian Splatting for 3D Head Avatar

arXiv 2507.18155 ↗First posted 24 Jul 2025cs.GR cs.CV cs.LGcited by 3

Not on the map. Why?

Abstract (arXiv)

Despite recent progress in 3D head avatar generation, balancing identity preservation, i.e., reconstruction, with novel poses and expressions, i.e., animation, remains a challenge. Existing methods struggle to adapt Gaussians to varying geometrical deviations across facial regions, resulting in suboptimal quality. To address this, we propose GeoAvatar, a framework for adaptive geometrical Gaussian Splatting. GeoAvatar leverages Adaptive Pre-allocation Stage (APS), an unsupervised method that segments Gaussians into rigid and flexible sets for adaptive offset regularization. Then, based on mouth anatomy and dynamics, we introduce a novel mouth structure and the part-wise deformation strategy to enhance the animation fidelity of the mouth. Finally, we propose a regularization loss for precise rigging between Gaussians and 3DMM faces. Moreover, we release DynamicFace, a video dataset with highly expressive facial motions. Extensive experiments show the superiority of GeoAvatar compared to state-of-the-art methods in reconstruction and novel animation scenarios.

Key facts. Cited by 3 papers in the map (data through 14 Oct 2026). Builds on 45 of its 64 references that are in the map. First posted on arXiv 24 Jul 2025.

Builds on

45 of 64 references.

References · 45

  1. 2017Attention Is All You Needcited by 60,361
  2. 2018The Unreasonable Effectiveness of Deep Features as a Perceptual Metriccited by 11,177
  3. 20233D Gaussian Splatting for Real-Time Radiance Field Renderingcited by 6,226
  4. 2022Instant Neural Graphics Primitives with a Multiresolution Hash Encodingcited by 3,723
  5. 2021Efficient Geometry-aware 3D Generative Adversarial Networkscited by 1,205
  6. 2020Nerfies: Deformable Neural Radiance Fieldscited by 1,198
  7. 20242D Gaussian Splatting for Geometrically Accurate Radiance Fieldscited by 1,076
  8. 2018BiSeNet: Bilateral Segmentation Network for Real-time Semantic Segmentationcited by 645
  9. 2020Dynamic Neural Radiance Fields for Monocular 4D Facial Avatar Reconstructioncited by 467
  10. 2019Capture, Learning, and Synthesis of 3D Speaking Stylescited by 259
  11. 2021I M Avatar: Implicit Morphable Head Avatars from Videoscited by 231
  12. 2021Neural Head Avatars from Monocular RGB Videoscited by 213
33 more
  1. 2023NeRSemble: Multi-view Radiance Field Reconstruction of Human Headscited by 194
  2. 2023COLMAP-Free 3D Gaussian Splattingcited by 193
  3. 2022PointAvatar: Deformable Point-based Head Avatars from Videoscited by 184
  4. 2020Real-Time High-Resolution Background Mattingcited by 173
  5. 2024SplattingAvatar: Realistic Real-Time Human Avatars with Mesh-Embedded Gaussian Splattingcited by 172
  6. 2022Towards Metrical Reconstruction of Human Facescited by 166
  7. 2022Instant Volumetric Head Avatarscited by 159
  8. 2017Morphable Face Models - An Open Frameworkcited by 138
  9. 2018DynaSLAM: Tracking, Mapping and Inpainting in Dynamic Scenescited by 134
  10. 2022Next3D: Generative Neural Texture Rasterization for 3D-Aware Head Avatarscited by 124
  11. 2024Generalizable and Animatable Gaussian Head Avatarcited by 97
  12. 2021NPMs: Neural Parametric Models for 3D Deformable Shapescited by 90
  13. 2023MonoGaussianAvatar: Monocular Gaussian Point-based Head Avatarcited by 89
  14. 2022AvatarMAV: Fast 3D Head Avatar Reconstruction Using Motion-Aware Neural Voxelscited by 73
  15. 2022Learning Neural Parametric Head Modelscited by 73
  16. 2023GAvatar: Animatable 3D Gaussian Avatars with Implicit Mesh Learningcited by 63
  17. 2023HeadGaS: Real-Time Animatable Head Avatars via 3D Gaussian Splattingcited by 60
  18. 2024TalkingGaussian: Structure-Persistent 3D Talking Head Synthesis via Gaussian Splattingcited by 60
  19. 20243D Gaussian Blendshapes for Head Avatar Animationcited by 58
  20. 2021Towards High Fidelity Monocular Face Reconstruction with Rich Reflectance using Self-supervised Learning and Ray Tracingcited by 48
  21. 2024GaussianHair: Hair Modeling and Rendering with Light-aware Gaussianscited by 48
  22. 2024HeadGAP: Few-Shot 3D Head Avatar via Generalizable Gaussian Priorscited by 44
  23. 2024MeGA: Hybrid Mesh-Gaussian Head Avatar for High-Fidelity Rendering and Head Editingcited by 38
  24. 2024DifFRelight: Diffusion-Based Facial Performance Relightingcited by 35
  25. 2023OmniAvatar: Geometry-Guided Controllable 3D Head Synthesiscited by 34
  26. 2024Gaussian Garments: Reconstructing Simulation-Ready Clothing with Photorealistic Appearance from Multi-View Videocited by 25
  27. 2020Survey on 3D face reconstruction from uncalibrated imagescited by 24
  28. 2023HiFace: High-Fidelity 3D Face Reconstruction by Learning Static and Dynamic Detailscited by 22
  29. 2024LayGA: Layered Gaussian Avatars for Animatable Clothing Transfercited by 22
  30. 2024GeneAvatar: Generic Expression-Aware Volumetric Head Avatar Editing from a Single Imagecited by 12
  31. 2024Correspondence-Guided SfM-Free 3D Gaussian Splatting for NVScited by 8
  32. 2024FAGhead: Fully Animate Gaussian Head from Monocular Videoscited by 2
  33. 2024Towards High-fidelity Head Blending with Chroma Keying for Industrial Applicationscited by 1

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2026: 3.

3 papers.

  1. 2026LentiAvatar: Pseudo-Multiview Reconstruction and Subpixel Prism Rendering for Real-Time Stereoscopic Communicationcited by 0
  2. 2026SpatialAvatar-0: High-Quality 4D Head Avatar with Multi-Stage Reconstructioncited by 0
  3. 2026S-Avatar: Diffusion-Guided Gaussian Head Avatars from a Single Imagecited by 0