mlsanityData to 19 Oct 2026

EGGHead: Efficient Generalizable Gaussian Heads from a Single Image

arXiv 2610.26898 ↗First posted 16 Oct 2026cs.CVcited by 0

On the map: 3D reconstruction and neural rendering → Avatars

Abstract (arXiv)

Creating photorealistic human heads typically involves trade-offs between representation power, efficiency, and generalizability. Traditional mesh-based solutions relying on 3DMMs generally sacrifice representation power but are efficient to render. On the other hand, NeRF-based methods improve representation power and fidelity while sacrificing efficiency. Gaussian splatting methods attempt to resolve these trade-offs by offering both high representation power and efficiency with real-time rendering. However, their efficiency nonetheless depends on the number of rasterized Gaussians, and the time needed to optimize and prepare the head (the enrollment time) is often high. In light of this, we propose EGGHead, which introduces a novel Gaussian allocation algorithm that efficiently assigns Gaussians to the head's semantic regions given a fixed Gaussian budget. Our algorithm reduces the number of Gaussians needed while minimizing the impact on representation power. EGGHead operates in a single forward pass and outperforms previous single-image face modeling methods in reconstruction quality, novel view synthesis, and rendering speeds.

Key facts. No citing papers in the map yet (data through 19 Oct 2026). Placed in Avatars, one of 15 directions in 3D reconstruction and neural rendering. Builds on 76 of its 98 references that are in the map. First posted on arXiv 16 Oct 2026.

Builds on

76 of 98 references.

Closest ancestors · 9

Same direction, newest first

  1. 2025TaoAvatar: Real-Time Lifelike Full-Body Talking Avatars for Augmented Reality via 3D Gaussian Splatting21 Mar
  2. 2024CAP4D: Creating Animatable 4D Portrait Avatars with Morphable Multi-View Diffusion Models16 Dec
  3. 2024GAF: Gaussian Avatar Reconstruction from Monocular Videos via Multi-view Diffusion13 Dec
  4. 2024Generalizable Human Gaussians for Sparse View Synthesis17 Jul
  5. 2024SplattingAvatar: Realistic Real-Time Human Avatars with Mesh-Embedded Gaussian Splatting8 Mar
  6. 20223DMM-RF: Convolutional Radiance Fields for 3D Face Modeling15 Sep
  7. 2022FDNeRF: Few-shot Dynamic Neural Radiance Fields for Face Reconstruction and Expression Editing11 Aug
  8. 2022RigNeRF: Fully Controllable Neural 3D Portraits13 Jun
  9. 2021MoFaNeRF: Morphable Facial Neural Radiance Field4 Dec

Other references · 67

  1. 2018The Unreasonable Effectiveness of Deep Features as a Perceptual Metriccited by 11,316
  2. 20233D Gaussian Splatting for Real-Time Radiance Field Renderingcited by 6,506
  3. 2018A Style-Based Generator Architecture for Generative Adversarial Networkscited by 6,036
  4. 2017Progressive Growing of GANs for Improved Quality, Stability, and Variationcited by 3,017
  5. 2020Nerfies: Deformable Neural Radiance Fieldscited by 1,217
  6. 2021Efficient Geometry-aware 3D Generative Adversarial Networkscited by 1,216
  7. 2020pi-GAN: Periodic Implicit Generative Adversarial Networks for 3D-Aware Image Synthesiscited by 713
  8. 2020GRAF: Generative Radiance Fields for 3D-Aware Image Synthesiscited by 697
  9. 2018BiSeNet: Bilateral Segmentation Network for Real-time Semantic Segmentationcited by 648
  10. 2023Towards Universal Fake Image Detectors that Generalize Across Generative Modelscited by 492
  11. 2021StyleNeRF: A Style-based 3D-Aware Generator for High-resolution Image Synthesiscited by 483
  12. 2020Learning an Animatable Detailed 3D Face Model from In-The-Wild Imagescited by 479
38 more
  1. 2020Dynamic Neural Radiance Fields for Monocular 4D Facial Avatar Reconstructioncited by 471
  2. 2023LightGaussian: Unbounded 3D Gaussian Compression with 15x Reduction and 200+ FPScited by 468
  3. 2022AdaFace: Quality Adaptive Margin for Face Recognitioncited by 395
  4. 2022On the detection of synthetic images generated by diffusion modelscited by 287
  5. 2023GaussianAvatars: Photorealistic Head Avatars with Rigged 3D Gaussianscited by 266
  6. 2023Compressed 3D Gaussian Splatting for Accelerated Novel View Synthesiscited by 255
  7. 2024Sapiens: Foundation for Human Vision Modelscited by 236
  8. 2021HeadNeRF: A Real-time NeRF-based Parametric Head Modelcited by 230
  9. 2017GANs Trained by a Two Time-Scale Update Rule Converge to a Local Nash Equilibriumcited by 227
  10. 2021GRAM: Generative Radiance Manifolds for 3D-Aware Image Generationcited by 217
  11. 2021Neural Head Avatars from Monocular RGB Videoscited by 214
  12. 2023EAGLES: Efficient Accelerated 3D Gaussians with Lightweight EncodingScited by 213
  13. 2023NeRSemble: Multi-view Radiance Field Reconstruction of Human Headscited by 196
  14. 2023GaussianAvatar: Towards Realistic Human Avatar Modeling from a Single Video via Animatable 3D Gaussianscited by 195
  15. 2023GPS-Gaussian: Generalizable Pixel-wise 3D Gaussian Splatting for Real-time Human Novel View Synthesiscited by 174
  16. 2018Nonlinear 3D Face Morphable Modelcited by 158
  17. 2024Reducing the Memory Footprint of 3D Gaussian Splattingcited by 149
  18. 2023CompGS: Smaller and Faster Gaussian Splatting with Vector Quantizationcited by 141
  19. 2023Hierarchical Fine-Grained Image Forgery Detection and Localizationcited by 135
  20. 2023Relightable Gaussian Codec Avatarscited by 130
  21. 2022FaceVerse: a Fine-grained and Detail-controllable 3D Face Morphable Model from a Hybrid Datasetcited by 99
  22. 2024Generalizable and Animatable Gaussian Head Avatarcited by 98
  23. 2023PanoHead: Geometry-Aware 3D Full-Head Synthesis in 360circcited by 95
  24. 2019Towards High-fidelity Nonlinear 3D Face Morphable Modelcited by 92
  25. 2023MonoGaussianAvatar: Monocular Gaussian Point-based Head Avatarcited by 91
  26. 2020Self-Supervised Monocular 3D Face Reconstruction by Occlusion-Aware Multi-view Geometry Consistencycited by 79
  27. 2020Learning Compositional Radiance Fields of Dynamic Human Headscited by 79
  28. 2023DiffusionRig: Learning Personalized Priors for Facial Appearance Editingcited by 76
  29. 2020Portrait Neural Radiance Fields from a Single Imagecited by 62
  30. 20243D Gaussian Blendshapes for Head Avatar Animationcited by 60
  31. 2021Reverse Engineering of Generative Models: Inferring Model Hyperparameters from Generated Imagescited by 59
  32. 2018On Learning 3D Face Morphable Model from In-the-wild Imagescited by 55
  33. 2024GGHead: Fast and Generalizable 3D Gaussian Headscited by 53
  34. 2020Towards High-Fidelity 3D Face Reconstruction from In-the-Wild Images Using Graph Convolutional Networkscited by 52
  35. 2022Face Relighting with Geometrically Consistent Shadowscited by 45
  36. 2023Detecting Images Generated by Deep Diffusion Models using their Local Intrinsic Dimensionalitycited by 45
  37. 2022FFHQ-UV: Normalized Facial UV-Texture Dataset for 3D Face Reconstructioncited by 40
  38. 2021Riggable 3D Face Reconstruction via In-Network Optimizationcited by 32

50 most cited of 67 All 67 (JSON)

Built on by

Not cited yet.

Similar, not cited

Close in wording, no citation either way

  1. 2024GaussianHeads: End-to-End Learning of Drivable Gaussian Head Avatars from Coarse-to-fine Representations
  2. 2023GaussianHead: High-fidelity Head Avatars with Learnable Gaussian Derivation
  3. 2026Split and Drive: Dual-Axis Disentanglement for Real-Time Gaussian Head Avatars
  4. 2023Gaussian3Diff: 3D Gaussian Diffusion for 3D Full Head Synthesis and Editing
  5. 2025SEGA: Drivable 3D Gaussian Head Avatar from a Single Image
  6. 2026SplitAvatar: One-shot Head Avatar with Autoregressive Gaussian Splitting
  7. 2025Using Gaussian Splats to Create High-Fidelity Facial Geometry and Texture
  8. 2026Feed-forward Gaussian Registration for Head Avatar Creation and Editing